Jakub Kicinski [Tue, 8 Apr 2025 19:59:52 +0000 (12:59 -0700)]
xdp: double protect netdev->xdp_flags with netdev->lock
Protect xdp_features with netdev->lock. This way pure readers
no longer have to take rtnl_lock to access the field.
This includes calling NETDEV_XDP_FEAT_CHANGE under the lock.
Looks like that's fine for bonding, the only "real" listener,
it's the same as ethtool feature change.
In terms of normal drivers - only GVE need special consideration
(other drivers don't use instance lock or don't support XDP).
It calls xdp_set_features_flag() helper from gve_init_priv() which
in turn is called from gve_reset_recovery() (locked), or prior
to netdev registration. So switch to _locked.
Reviewed-by: Joe Damato <jdamato@fastly.com> Acked-by: Stanislav Fomichev <sdf@fomichev.me> Acked-by: Harshitha Ramamurthy <hramamurthy@google.com> Acked-by: Martin KaFai Lau <martin.lau@kernel.org> Link: https://patch.msgid.link/20250408195956.412733-6-kuba@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski [Tue, 8 Apr 2025 19:59:51 +0000 (12:59 -0700)]
netdev: don't hold rtnl_lock over nl queue info get when possible
Netdev queue dump accesses: NAPI, memory providers, XSk pointers.
All three are "ops protected" now, switch to the op compat locking.
rtnl lock does not have to be taken for "ops locked" devices.
Jakub Kicinski [Tue, 8 Apr 2025 19:59:50 +0000 (12:59 -0700)]
netdev: add "ops compat locking" helpers
Add helpers to "lock a netdev in a backward-compatible way",
which for ops-locked netdevs will mean take the instance lock.
For drivers which haven't opted into the ops locking we'll take
rtnl_lock.
The scoped foreach is dropping and re-taking the lock for each
device, even if prev and next are both under rtnl_lock.
I hope that's fine since we expect that netdev nl to be mostly
supported by modern drivers, and modern drivers should also
opt into the instance locking.
Note that these helpers are mostly needed for queue related state,
because drivers modify queue config in their ops in a non-atomic
way. Or differently put, queue changes don't have a clear-cut API
like NAPI configuration. Any state that can should just use the
instance lock directly, not the "compat" hacks.
Jakub Kicinski [Tue, 8 Apr 2025 19:59:49 +0000 (12:59 -0700)]
net: designate XSK pool pointers in queues as "ops protected"
Read accesses go via xsk_get_pool_from_qid(), the call coming
from the core and gve look safe (other "ops locked" drivers
don't support XSK).
Write accesses go via xsk_reg_pool_at_qid() and xsk_clear_pool_at_qid().
Former is already under the ops lock, latter is not (both coming from
the workqueue via xp_clear_dev() and NETDEV_UNREGISTER via xsk_notifier()).
Jakub Kicinski [Tue, 8 Apr 2025 19:59:48 +0000 (12:59 -0700)]
net: avoid potential race between netdev_get_by_index_lock() and netns switch
netdev_get_by_index_lock() performs following steps:
rcu_lock();
dev = lookup(netns, ifindex);
dev_get(dev);
rcu_unlock();
[... lock & validate the dev ...]
return dev
Validation right now only checks if the device is registered but since
the lookup is netns-aware we must also protect against the device
switching netns right after we dropped the RCU lock. Otherwise
the caller in netns1 may get a pointer to a device which has just
switched to netns2.
We can't hold the lock for the entire netns change process (because of
the NETDEV_UNREGISTER notifier), and there's no existing marking to
indicate that the netns is unlisted because of netns move, so add one.
AFAIU none of the existing netdev_get_by_index_lock() callers can
suffer from this problem (NAPI code double checks the netns membership
and other callers are either under rtnl_lock or not ns-sensitive),
so this patch does not have to be treated as a fix.
Michal Luczaj [Mon, 7 Apr 2025 19:01:02 +0000 (21:01 +0200)]
net: Drop unused @sk of __skb_try_recv_from_queue()
__skb_try_recv_from_queue() deals with a queue, @sk is not used
since commit e427cad6eee4 ("net: datagram: drop 'destructor'
argument from several helpers"). Remove sk from function parameters,
adapt callers.
The UDP tunnel GRO stage is source of measurable overhead for workload
based on UDP-encapsulated traffic: each incoming packets requires a full
UDP socket lookup and an indirect call.
In the most common setups a single UDP tunnel device is used. In such
case we can optimize both the lookup and the indirect call.
Patch 1 tracks per netns the active UDP tunnels and replaces the socket
lookup with a single destination port comparison when possible.
Patch 2 tracks the different types of UDP tunnels and replaces the
indirect call with a static one when there is a single UDP tunnel type
active.
I measure ~10% performance improvement in TCP over UDP tunnel stream
tests on top of this series.
Paolo Abeni [Mon, 7 Apr 2025 15:45:42 +0000 (17:45 +0200)]
udp_tunnel: use static call for GRO hooks when possible
It's quite common to have a single UDP tunnel type active in the
whole system. In such a case we can replace the indirect call for
the UDP tunnel GRO callback with a static call.
Add the related accounting in the control path and switch to static
call when possible. To keep the code simple use a static array for
the registered tunnel types, and size such array based on the kernel
config.
Note that there are valid kernel configurations leading to
UDP_MAX_TUNNEL_TYPES == 0 even with IS_ENABLED(CONFIG_NET_UDP_TUNNEL),
Explicitly skip the accounting in such a case, to avoid compile warning
when accessing "udp_tunnel_gro_types".
Paolo Abeni [Mon, 7 Apr 2025 15:45:41 +0000 (17:45 +0200)]
udp_tunnel: create a fastpath GRO lookup.
Most UDP tunnels bind a socket to a local port, with ANY address, no
peer and no interface index specified.
Additionally it's quite common to have a single tunnel device per
namespace.
Track in each namespace the UDP tunnel socket respecting the above.
When only a single one is present, store a reference in the netns.
When such reference is not NULL, UDP tunnel GRO lookup just need to
match the incoming packet destination port vs the socket local port.
The tunnel socket never sets the reuse[port] flag[s]. When bound to no
address and interface, no other socket can exist in the same netns
matching the specified local port.
Matching packets with non-local destination addresses will be
aggregated, and eventually segmented as needed - no behavior changes
intended.
Restrict the optimization to kernel sockets only: it covers all the
relevant use-cases, and user-space owned sockets could be disconnected
and rebound after setup_udp_tunnel_sock(), breaking the uniqueness
assumption
Note that the UDP tunnel socket reference is stored into struct
netns_ipv4 for both IPv4 and IPv6 tunnels. That is intentional to keep
all the fastpath-related netns fields in the same struct and allow
cacheline-based optimization. Currently both the IPv4 and IPv6 socket
pointer share the same cacheline as the `udp_table` field.
Victor Nogueira [Mon, 7 Apr 2025 21:56:56 +0000 (18:56 -0300)]
selftests: tc-testing: Pre-load IFE action and its submodules
Recently we had some issues in parallel TDC where some of IFE tests are
failing due to some of IFE's submodules (like act_meta_skbtcindex and
act_meta_skbprio) taking too long to load [1]. To avoid that issue,
pre-load IFE and all its submodules before running any of the tests in
tdc.sh
Eric Dumazet [Mon, 7 Apr 2025 16:35:59 +0000 (16:35 +0000)]
net: rps: change skb_flow_limit() hash function
As explained in commit f3483c8e1da6 ("net: rfs: hash function change"),
masking low order bits of skb_get_hash(skb) has low entropy.
A NIC with 32 RX queues uses the 5 low order bits of rss key
to select a queue. This means all packets landing to a given
queue share the same 5 low order bits.
Use SPDX-License-Identifier accross all the files of the xgbe driver to
ensure compliance with Linux kernel standards, thus removing the
boiler-plate template license text.
If the destination buffer has a fixed length, strscpy_pad()
automatically determines its size using sizeof() when the argument is
omitted. This makes the explicit sizeof() calls unnecessary - remove
them.
Merge tag 'net-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Jakub Kicinski:
"Including fixes from netfilter.
Current release - regressions:
- four fixes for the netdev per-instance locking
Current release - new code bugs:
- consolidate more code between existing Rx zero-copy and uring so
that the latter doesn't miss / have to duplicate the safety checks
Previous releases - regressions:
- ipv6: fix omitted Netlink attributes when using SKIP_STATS
Previous releases - always broken:
- net: fix geneve_opt length integer overflow
- udp: fix multiple wrap arounds of sk->sk_rmem_alloc when it
approaches INT_MAX
- dsa: mvpp2: add a lock to avoid corruption of the shared TCAM
- dsa: airoha: fix issues with traffic QoS configuration / offload,
and flow table offload
Misc:
- touch up the Netlink YAML specs of old families to make them usable
for user space C codegen"
* tag 'net-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (56 commits)
selftests: net: amt: indicate progress in the stress test
netlink: specs: rt_route: pull the ifa- prefix out of the names
netlink: specs: rt_addr: pull the ifa- prefix out of the names
netlink: specs: rt_addr: fix get multi command name
netlink: specs: rt_addr: fix the spec format / schema failures
net: avoid false positive warnings in __net_mp_close_rxq()
net: move mp dev config validation to __net_mp_open_rxq()
net: ibmveth: make veth_pool_store stop hanging
arcnet: Add NULL check in com20020pci_probe()
ipv6: Do not consider link down nexthops in path selection
ipv6: Start path selection from the first nexthop
usbnet:fix NPE during rx_complete
net: octeontx2: Handle XDP_ABORTED and XDP invalid as XDP_DROP
net: fix geneve_opt length integer overflow
io_uring/zcrx: fix selftests w/ updated netdev Python helpers
selftests: net: use netdevsim in netns test
docs: net: document netdev notifier expectations
net: dummy: request ops lock
netdevsim: add dummy device notifiers
net: rename rtnl_net_debug to lock_debug
...
Merge tag 'spi-fix-v6.15-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Pull spi fixes from Mark Brown:
"A small collection of fixes that came in during the merge window,
everything is driver specific with nothing standing out particularly"
* tag 'spi-fix-v6.15-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: bcm2835: Restore native CS probing when pinctrl-bcm2835 is absent
spi: bcm2835: Do not call gpiod_put() on invalid descriptor
spi: cadence-qspi: revert "Improve spi memory performance"
spi: cadence: Fix out-of-bounds array access in cdns_mrvl_xspi_setup_clock()
spi: fsl-qspi: use devm function instead of driver remove
spi: SPI_QPIC_SNAND should be tristate and depend on MTD
spi-rockchip: Fix register out of bounds access
Merge tag 'soc-drivers-6.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull more SoC driver updates from Arnd Bergmann:
"This is the promised follow-up to the soc drivers branch, adding minor
updates to omap and freescale drivers.
Most notably, Ioana Ciornei takes over maintenance of the DPAA bus
driver used in some NXP (originally Freescale) chips"
* tag 'soc-drivers-6.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
bus: fsl-mc: Remove deadcode
MAINTAINERS: add the linuppc-dev list to the fsl-mc bus entry
MAINTAINERS: fix nonexistent dtbinding file name
MAINTAINERS: add myself as maintainer for the fsl-mc bus
irqdomain: soc: Switch to irq_find_mapping()
Input: tsc2007 - accept standard properties
Merge tag 'platform-drivers-x86-v6.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver fixes from Ilpo Järvinen:
- thinkpad_acpi:
- Fix NULL pointer dereferences while probing
- Disable ACPI fan access for T495* and E560
- ISST: Correct command storage data length
* tag 'platform-drivers-x86-v6.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
MAINTAINERS: consistently use my dedicated email address
platform/x86: ISST: Correct command storage data length
platform/x86: thinkpad_acpi: disable ACPI fan access for T495* and E560
platform/x86: thinkpad_acpi: Fix NULL pointer dereferences while probing
Jakub Kicinski [Thu, 3 Apr 2025 14:56:36 +0000 (07:56 -0700)]
selftests: net: amt: indicate progress in the stress test
Our CI expects output from the test at least once every 10 minutes.
The AMT test when running on debug kernel is just on the edge
of that time for the stress test. Improve the output:
- print the name of the test first, before starting it,
- output a dot every 10% of the way.
Output after:
TEST: amt discovery [ OK ]
TEST: IPv4 amt multicast forwarding [ OK ]
TEST: IPv6 amt multicast forwarding [ OK ]
TEST: IPv4 amt traffic forwarding torture .......... [ OK ]
TEST: IPv6 amt traffic forwarding torture .......... [ OK ]
Jakub Kicinski [Thu, 3 Apr 2025 01:37:06 +0000 (18:37 -0700)]
netlink: specs: rt_route: pull the ifa- prefix out of the names
YAML specs don't normally include the C prefix name in the name
of the YAML attr. Remove the ifa- prefix from all attributes
in route-attrs and metrics and specify name-prefix instead.
This is a bit risky, hopefully there aren't many users out there.
Jakub Kicinski [Thu, 3 Apr 2025 01:37:05 +0000 (18:37 -0700)]
netlink: specs: rt_addr: pull the ifa- prefix out of the names
YAML specs don't normally include the C prefix name in the name
of the YAML attr. Remove the ifa- prefix from all attributes
in addr-attrs and specify name-prefix instead.
This is a bit risky, hopefully there aren't many users out there.
====================
net: make memory provider install / close paths more common
We seem to be fixing bugs in config path for devmem which also exist
in the io_uring ZC path. Let's try to make the two paths more common,
otherwise this is bound to keep happening.
Jakub Kicinski [Thu, 3 Apr 2025 01:34:05 +0000 (18:34 -0700)]
net: avoid false positive warnings in __net_mp_close_rxq()
Commit under Fixes solved the problem of spurious warnings when we
uninstall an MP from a device while its down. The __net_mp_close_rxq()
which is used by io_uring was not fixed. Move the fix over and reuse
__net_mp_close_rxq() in the devmem path.
Acked-by: Stanislav Fomichev <sdf@fomichev.me> Fixes: a70f891e0fa0 ("net: devmem: do not WARN conditionally after netdev_rx_queue_restart()") Reviewed-by: Mina Almasry <almasrymina@google.com> Link: https://patch.msgid.link/20250403013405.2827250-3-kuba@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski [Thu, 3 Apr 2025 01:34:04 +0000 (18:34 -0700)]
net: move mp dev config validation to __net_mp_open_rxq()
devmem code performs a number of safety checks to avoid having
to reimplement all of them in the drivers. Move those to
__net_mp_open_rxq() and reuse that function for binding to make
sure that io_uring ZC also benefits from them.
While at it rename the queue ID variable to rxq_idx in
__net_mp_open_rxq(), we touch most of the relevant lines.
The XArray insertion is reordered after the netdev_rx_queue_restart()
call, otherwise we'd need to duplicate the queue index check
or risk inserting an invalid pointer. The XArray allocation
failures should be extremely rare.
Reviewed-by: Mina Almasry <almasrymina@google.com> Acked-by: Stanislav Fomichev <sdf@fomichev.me> Fixes: 6e18ed929d3b ("net: add helpers for setting a memory provider on an rx queue") Link: https://patch.msgid.link/20250403013405.2827250-2-kuba@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Dave Marquardt [Wed, 2 Apr 2025 15:44:03 +0000 (10:44 -0500)]
net: ibmveth: make veth_pool_store stop hanging
v2:
- Created a single error handling unlock and exit in veth_pool_store
- Greatly expanded commit message with previous explanatory-only text
Summary: Use rtnl_mutex to synchronize veth_pool_store with itself,
ibmveth_close and ibmveth_open, preventing multiple calls in a row to
napi_disable.
Background: Two (or more) threads could call veth_pool_store through
writing to /sys/devices/vio/30000002/pool*/*. You can do this easily
with a little shell script. This causes a hang.
I configured LOCKDEP, compiled ibmveth.c with DEBUG, and built a new
kernel. I ran this test again and saw:
The control APIs are not idempotent. Control API calls are safe
against concurrent use of datapath APIs but an incorrect sequence of
control API calls may result in crashes, deadlocks, or race
conditions. For example, calling napi_disable() multiple times in a
row will deadlock.
In the normal open and close paths, rtnl_mutex is acquired to prevent
other callers. This is missing from veth_pool_store. Use rtnl_mutex in
veth_pool_store fixes these hangs.
Signed-off-by: Dave Marquardt <davemarq@linux.ibm.com> Fixes: 860f242eb534 ("[PATCH] ibmveth change buffer pools dynamically") Reviewed-by: Nick Child <nnac123@linux.ibm.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250402154403.386744-1-davemarq@linux.ibm.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Henry Martin [Wed, 2 Apr 2025 13:50:36 +0000 (21:50 +0800)]
arcnet: Add NULL check in com20020pci_probe()
devm_kasprintf() returns NULL when memory allocation fails. Currently,
com20020pci_probe() does not check for this case, which results in a
NULL pointer dereference.
Add NULL check after devm_kasprintf() to prevent this issue and ensure
no resources are left allocated.
ipv6: Do not consider link down nexthops in path selection
Nexthops whose link is down are not supposed to be considered during
path selection when the "ignore_routes_with_linkdown" sysctl is set.
This is done by assigning them a negative region boundary.
However, when comparing the computed hash (unsigned) with the region
boundary (signed), the negative region boundary is treated as unsigned,
resulting in incorrect nexthop selection.
Fix by treating the computed hash as signed. Note that the computed hash
is always in range of [0, 2^31 - 1].
Fixes: 3d709f69a3e7 ("ipv6: Use hash-threshold instead of modulo-N") Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Link: https://patch.msgid.link/20250402114224.293392-3-idosch@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cited commit transitioned IPv6 path selection to use hash-threshold
instead of modulo-N. With hash-threshold, each nexthop is assigned a
region boundary in the multipath hash function's output space and a
nexthop is chosen if the calculated hash is smaller than the nexthop's
region boundary.
Hash-threshold does not work correctly if path selection does not start
with the first nexthop. For example, if fib6_select_path() is always
passed the last nexthop in the group, then it will always be chosen
because its region boundary covers the entire hash function's output
space.
Fix this by starting the selection process from the first nexthop and do
not consider nexthops for which rt6_score_route() provided a negative
score.
Fixes: 3d709f69a3e7 ("ipv6: Use hash-threshold instead of modulo-N") Reported-by: Stanislav Fomichev <stfomichev@gmail.com> Closes: https://lore.kernel.org/netdev/Z9RIyKZDNoka53EO@mini-arch/ Signed-off-by: Ido Schimmel <idosch@nvidia.com> Link: https://patch.msgid.link/20250402114224.293392-2-idosch@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Ying Lu [Wed, 2 Apr 2025 08:58:59 +0000 (16:58 +0800)]
usbnet:fix NPE during rx_complete
Missing usbnet_going_away Check in Critical Path.
The usb_submit_urb function lacks a usbnet_going_away
validation, whereas __usbnet_queue_skb includes this check.
This inconsistency creates a race condition where:
A URB request may succeed, but the corresponding SKB data
fails to be queued.
Subsequent processes:
(e.g., rx_complete → defer_bh → __skb_unlink(skb, list))
attempt to access skb->next, triggering a NULL pointer
dereference (Kernel Panic).
net: octeontx2: Handle XDP_ABORTED and XDP invalid as XDP_DROP
In the current implementation octeontx2 manages XDP_ABORTED and XDP
invalid as XDP_PASS forwarding the skb to the networking stack.
Align the behaviour to other XDP drivers handling XDP_ABORTED and XDP
invalid as XDP_DROP.
Please note this patch has just compile tested.
Merge tag 'x86-urgent-2025-04-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
- Fix a performance regression on AMD iGPU and dGPU drivers, related to
the unintended activation of DMA bounce buffers that regressed game
performance if KASLR disturbed things just enough
- Fix a copy_user_generic() performance regression on certain older
non-FSRM/ERMS CPUs
- Fix a Clang build warning due to a semantic merge conflict the Kunit
tree generated with the x86 tree
- Fix FRED related system hang during S4 resume
- Remove an unused API
* tag 'x86-urgent-2025-04-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/fred: Fix system hang during S4 resume with FRED enabled
x86/platform/iosf_mbi: Remove unused iosf_mbi_unregister_pmic_bus_access_notifier()
x86/mm/init: Handle the special case of device private pages in add_pages(), to not increase max_pfn and trigger dma_addressing_limited() bounce buffers
x86/tools: Drop duplicate unlikely() definition in insn_decoder_test.c
x86/uaccess: Improve performance by aligning writes to 8 bytes in copy_user_generic(), on non-FSRM/ERMS CPUs
Merge tag 'sound-fix-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"A collection of device-specific fixes that have been gathered since
the previous pull:
- A few more HD-audio quirks and fixups
- A series of Qualcomm AudioReach fixes
- Various small fixes for ASoC rt5665, WSA, SOF and Cirrus"
* tag 'sound-fix-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda/realtek: Fix built-in mic on another ASUS VivoBook model
ALSA: hda/realtek - Support mute led function for HP platform
ASoC: imx-card: Add NULL check in imx_card_probe()
ASoC: codecs: rt5665: Fix some error handling paths in rt5665_probe()
ASoC: q6apm-dai: make use of q6apm_get_hw_pointer
ASoC: qdsp6: q6apm-dai: fix capture pipeline overruns.
ASoC: qdsp6: q6apm-dai: set 10 ms period and buffer alignment.
ASoC: q6apm: add q6apm_get_hw_pointer helper
ASoC: q6apm-dai: schedule all available frames to avoid dsp under-runs
ASoC: SOF: hda/ptl: Move mic privacy change notification sending to a work
ALSA/hda: intel-sdw-acpi: Remove (explicitly) unused header
ALSA: hda/realtek: Enable Mute LED on HP OMEN 16 Laptop xd000xx
ALSA: hda/tas2781: Upgrade calibratd-data writing code to support Alpha and Beta dsp firmware
ASoC: qdsp6: q6asm-dai: fix q6asm_dai_compr_set_params error path
ALSA: hda/realtek: Fix built-in mic breakage on ASUS VivoBook X515JA
ASoC: sma1307: Fix error handling in sma1307_setting_loaded()
ASoC: codecs: wsa884x: Correct VI sense channel mask
ASoC: codecs: wsa883x: Correct VI sense channel mask
firmware: cs_dsp: Ensure cs_dsp_load[_coeff]() returns 0 on success
Merge tag 'omap-for-v6.14/drivers-signed' of https://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-omap into soc/drivers-2
arm/omap: drivers: updates for v6.14
* tag 'omap-for-v6.14/drivers-signed' of https://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-omap:
Input: tsc2007 - accept standard properties
Merge tag 'soc_fsl-6.15-1' of https://github.com/chleroy/linux into soc/drivers-2
FSL SOC Changes for 6.15:
- irqdomain cleanups from Jiry
- Add Ioana as Maintainer of fsl-mc bus and remove Laurentiu and Stuart
- Remove deadcode from fsl-mc bus
* tag 'soc_fsl-6.15-1' of https://github.com/chleroy/linux:
bus: fsl-mc: Remove deadcode
MAINTAINERS: add the linuppc-dev list to the fsl-mc bus entry
MAINTAINERS: fix nonexistent dtbinding file name
MAINTAINERS: add myself as maintainer for the fsl-mc bus
irqdomain: soc: Switch to irq_find_mapping()
Merge tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull dcache fixes from Al Viro:
"Fixes for bugs caught as part of tree-in-dcache work.
Mostly dentry refcount mishandling"
* tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
hypfs_create_cpu_files(): add missing check for hypfs_mkdir() failure
qibfs: fix _another_ leak
spufs: fix a leak in spufs_create_context()
spufs: fix gang directory lifetimes
spufs: fix a leak on spufs_new_file() failure
Jakub Kicinski [Thu, 3 Apr 2025 23:23:00 +0000 (16:23 -0700)]
Merge tag 'nf-25-04-03' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following batch contains Netfilter fixes for net:
1) conncount incorrectly removes element for non-dynamic sets,
these elements represent a static control plane configuration,
leave them in place.
2) syzbot found a way to unregister a basechain that has been never
registered from the chain update path, fix from Florian Westphal.
3) Fix incorrect pointer arithmetics in geneve support for tunnel,
from Lin Ma.
* tag 'nf-25-04-03' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
netfilter: nft_tunnel: fix geneve_opt type confusion addition
netfilter: nf_tables: don't unregister hook when table is dormant
netfilter: nft_set_hash: GC reaps elements with conncount for dynamic sets only
====================
Merge tag 'v6.15rc-part2-ksmbd-server-fixes' of git://git.samba.org/ksmbd
Pull smb server fixes from Steve French:
"Four ksmbd SMB3 server fixes, all also for stable"
* tag 'v6.15rc-part2-ksmbd-server-fixes' of git://git.samba.org/ksmbd:
ksmbd: fix null pointer dereference in alloc_preauth_hash()
ksmbd: validate zero num_subauth before sub_auth is accessed
ksmbd: fix overflow in dacloffset bounds check
ksmbd: fix session use-after-free in multichannel connection
Merge tag 'trace-ringbuffer-v6.15-3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull ring-buffer updates from Steven Rostedt:
"Persistent buffer cleanups and simplifications.
It was mistaken that the physical memory returned from "reserve_mem"
had to be vmap()'d to get to it from a virtual address. But
reserve_mem already maps the memory to the virtual address of the
kernel so a simple phys_to_virt() can be used to get to the virtual
address from the physical memory returned by "reserve_mem". With this
new found knowledge, the code can be cleaned up and simplified.
- Enforce that the persistent memory is page aligned
As the buffers using the persistent memory are all going to be
mapped via pages, make sure that the memory given to the tracing
infrastructure is page aligned. If it is not, it will print a
warning and fail to map the buffer.
- Use phys_to_virt() to get the virtual address from reserve_mem
Instead of calling vmap() on the physical memory returned from
"reserve_mem", use phys_to_virt() instead.
As the memory returned by "memmap" or any other means where a
physical address is given to the tracing infrastructure, it still
needs to be vmap(). Since this memory can never be returned back to
the buddy allocator nor should it ever be memmory mapped to user
space, flag this buffer and up the ref count. The ref count will
keep it from ever being freed, and the flag will prevent it from
ever being memory mapped to user space.
- Use vmap_page_range() for memmap virtual address mapping
For the memmap buffer, instead of allocating an array of struct
pages, assigning them to the contiguous phsycial memory and then
passing that to vmap(), use vmap_page_range() instead
- Replace flush_dcache_folio() with flush_kernel_vmap_range()
Instead of calling virt_to_folio() and passing that to
flush_dcache_folio(), just call flush_kernel_vmap_range() directly.
This also fixes a bug where if a subbuffer was bigger than
PAGE_SIZE only the PAGE_SIZE portion would be flushed"
* tag 'trace-ringbuffer-v6.15-3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
ring-buffer: Use flush_kernel_vmap_range() over flush_dcache_folio()
tracing: Use vmap_page_range() to map memmap ring buffer
tracing: Have reserve_mem use phys_to_virt() and separate from memmap buffer
tracing: Enforce the persistent ring buffer to be page aligned
Merge tag 'block-6.15-20250403' of git://git.kernel.dk/linux
Pull more block updates from Jens Axboe:
- NVMe pull request via Keith:
- PCI endpoint target cleanup (Damien)
- Early import for uring_cmd fixed buffer (Caleb)
- Multipath documentation and notification improvements (John)
- Invalid pci sq doorbell write fix (Maurizio)
- Queue init locking fix
- Remove dead nsegs parameter from blk_mq_get_new_requests()
* tag 'block-6.15-20250403' of git://git.kernel.dk/linux:
block: don't grab elevator lock during queue initialization
nvme-pci: skip nvme_write_sq_db on empty rqlist
nvme-multipath: change the NVME_MULTIPATH config option
nvme: update the multipath warning in nvme_init_ns_head
nvme/ioctl: move fixed buffer lookup to nvme_uring_cmd_io()
nvme/ioctl: move blk_mq_free_request() out of nvme_map_user_request()
nvme/ioctl: don't warn on vectorized uring_cmd with fixed buffer
nvmet: pci-epf: Keep completion queues mapped
block: remove unused nseg parameter
For igc:
Joe Damato removes unmapping of XSK queues from NAPI instance.
Zdenek Bouska swaps condition checks/call to prevent AF_XDP Tx drops
with low budget value.
For e1000e:
Vitaly adjusts Kumeran interface configuration to prevent MDI errors.
For ixgbe:
Piotr clears PHY high values on media type detection to ensure stale
values are not used.
For idpf:
Emil adjusts shutdown calls to prevent NULL pointer dereference.
* '1GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
idpf: fix adapter NULL pointer dereference on reboot
ixgbe: fix media type detection for E610 device
e1000e: change k1 configuration on MTP and later platforms
igc: Fix TX drops in XDP ZC
igc: Fix XSK queue NAPI ID mapping
====================
Merge tag 'io_uring-6.15-20250403' of git://git.kernel.dk/linux
Pull more io_uring updates from Jens Axboe:
"Set of fixes/updates for io_uring that should go into this release.
The ublk bits could've gone via either tree - usually I put them in
block, but they got a bit mixed this series with the zero-copy
supported that ended up dipping into both trees.
This contains:
- Fix for sendmsg zc, include in pinned pages accounting like we do
for the other zc types
- Series for ublk fixing request aborting, doing various little
cleanups, fixing some zc issues, and adding queue_rqs support
- Another ublk series doing some code cleanups
- Series cleaning up the io_uring send path, mostly in preparation
for registered buffers
- Series doing little MSG_RING cleanups
- Fix for the newly added zc rx, fixing len being 0 for the last
invocation of the callback
- Add vectored registered buffer support for ublk. With that, then
ublk also supports this feature in the kernel revision where it
could generically introduced for rw/net
- A bunch of selftest additions for ublk. This is the majority of the
diffstat
- Silence a KCSAN data race warning for io-wq
- Various little cleanups and fixes"
* tag 'io_uring-6.15-20250403' of git://git.kernel.dk/linux: (44 commits)
io_uring: always do atomic put from iowq
selftests: ublk: enable zero copy for stripe target
io_uring: support vectored kernel fixed buffer
block: add for_each_mp_bvec()
io_uring: add validate_fixed_range() for validate fixed buffer
selftests: ublk: kublk: fix an error log line
selftests: ublk: kublk: use ioctl-encoded opcodes
io_uring/zcrx: return early from io_zcrx_recv_skb if readlen is 0
io_uring/net: avoid import_ubuf for regvec send
io_uring/rsrc: check size when importing reg buffer
io_uring: cleanup {g,s]etsockopt sqe reading
io_uring: hide caches sqes from drivers
io_uring: make zcrx depend on CONFIG_IO_URING
io_uring: add req flag invariant build assertion
Documentation: ublk: remove dead footnote
selftests: ublk: specify io_cmd_buf pointer type
ublk: specify io_cmd_buf pointer type
io_uring: don't pass ctx to tw add remote helper
io_uring/msg: initialise msg request opcode
io_uring/msg: rename io_double_lock_ctx()
...
Lin Ma [Wed, 2 Apr 2025 16:56:32 +0000 (00:56 +0800)]
net: fix geneve_opt length integer overflow
struct geneve_opt uses 5 bit length for each single option, which
means every vary size option should be smaller than 128 bytes.
However, all current related Netlink policies cannot promise this
length condition and the attacker can exploit a exact 128-byte size
option to *fake* a zero length option and confuse the parsing logic,
further achieve heap out-of-bounds read.
Fix these issues by enforing correct length condition in related
policies.
Fixes: 925d844696d9 ("netfilter: nft_tunnel: add support for geneve opts") Fixes: 4ece47787077 ("lwtunnel: add options setting and dumping for geneve") Fixes: 0ed5269f9e41 ("net/sched: add tunnel option support to act_tunnel_key") Fixes: 0a6e77784f49 ("net/sched: allow flower to match tunnel options") Signed-off-by: Lin Ma <linma@zju.edu.cn> Reviewed-by: Xin Long <lucien.xin@gmail.com> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Link: https://patch.msgid.link/20250402165632.6958-1-linma@zju.edu.cn Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Merge tag '9p-for-6.15-rc1' of https://github.com/martinetd/linux
Pull 9p updates from Dominique Martinet:
- fix handling of bogus (negative/too long) replies
- fix crash on mkdir with ACLs (... looks like nobody is using ACLs
with semi-recent kernels...)
- ipv6 support for trans=tcp
- minor concurrency fix to make syzbot happy
- minor cleanup
* tag '9p-for-6.15-rc1' of https://github.com/martinetd/linux:
docs: fs/9p: Add missing "not" in cache documentation
9p: Use hashtable.h for hash_errmap
Documentation/fs/9p: fix broken link
9p/trans_fd: mark concurrent read and writes to p9_conn->err
9p/net: return error on bogus (longer than requested) replies
9p/net: fix improper handling of bogus negative read/write replies
fs/9p: fix NULL pointer dereference on mkdir
net/9p/fd: support ipv6 for trans=tcp
====================
net: hold instance lock during NETDEV_UP/REGISTER
Solving the issue reported by Cosmin in [0] requires consistent
lock during NETDEV_UP/REGISTER notifiers. This series
addresses that (along with some other fixes in net/ipv4/devinet.c
and net/ipv6/addrconf.c) and appends the patches from Jakub
that were conditional on consistent locking in NETDEV_UNREGISTER.
Even though dummy device doesn't really need an instance lock,
a lot of selftests use dummy so it's useful to have extra
expose to the instance lock on NIPA. Request the instance/ops
locking.
In order to exercise and verify notifiers' locking assumptions,
register dummy notifiers (via register_netdevice_notifier_dev_net).
Share notifier event handler that enforces the assumptions with
lock_debug.c (rename and export rtnl_net_debug_event as
netdev_debug_event). Add ops lock asserts to netdev_debug_event.
ipv6_add_dev might call dev_disable_lro which unconditionally grabs
instance lock, so it will deadlock during NETDEV_REGISTER. Switch
to netif_disable_lro.
Make sure all callers hold the instance lock as well.
Cc: Cosmin Ratiu <cratiu@nvidia.com> Fixes: ad7c7b2172c3 ("net: hold netdev instance lock during sysfs operations") Signed-off-by: Stanislav Fomichev <sdf@fomichev.me> Link: https://patch.msgid.link/20250401163452.622454-4-sdf@fomichev.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Callers of inetdev_init can come from several places with inconsistent
expectation about netdev instance lock. Grab instance lock during
REGISTER (plus UP). Also solve the inconsistency with UNREGISTER
where it was locked only during move netns path.
Switch to netif_disable_lro which assumes the caller holds the instance
lock. inetdev_init is called for blackhole device (which sw device and
doesn't grab instance lock) and from REGISTER/UNREGISTER notifiers.
We already hold the instance lock for REGISTER notifier during
netns change and we'll soon hold the lock during other paths.
Reviewed-by: Jakub Kicinski <kuba@kernel.org> Reported-by: Cosmin Ratiu <cratiu@nvidia.com> Fixes: ad7c7b2172c3 ("net: hold netdev instance lock during sysfs operations") Signed-off-by: Stanislav Fomichev <sdf@fomichev.me> Link: https://patch.msgid.link/20250401163452.622454-2-sdf@fomichev.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Merge tag 'rtc-6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux
Pull RTC updates from Alexandre Belloni:
"We see a net reduction of the number of lines of code thanks to the
removal of a now unused driver and a testing tool that is not used
anymore. Apart from this, the max31335 driver gets support for a new
part number and pm8xxx gets UEFI support.
Core:
- setdate is removed as it has better replacements
- skip alarms with a second resolution when we know the RTC doesn't
support those.
Subsystem:
- remove unnecessary private struct members
- use devm_pm_set_wake_irq were relevant
Drivers:
- ds1307: stop disabling alarms on probe for DS1337, DS1339, DS1341
and DS3231
- max31335: add max31331 support
- pcf50633 is removed as support for the related SoC has been removed
- pcf85063: properly handle POR failures"
* tag 'rtc-6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux: (50 commits)
rtc: remove 'setdate' test program
selftest: rtc: skip some tests if the alarm only supports minutes
rtc: mt6397: drop unused defines
rtc: pcf85063: replace dev_err+return with return dev_err_probe
rtc: pcf85063: do a SW reset if POR failed
rtc: max31335: Add driver support for max31331
dt-bindings: rtc: max31335: Add max31331 support
rtc: cros-ec: Avoid a couple of -Wflex-array-member-not-at-end warnings
dt-bindings: rtc: pcf2127: Reference spi-peripheral-props.yaml
rtc: rzn1: implement one-second accuracy for alarms
rtc: pcf50633: Remove
rtc: pm8xxx: implement qcom,no-alarm flag for non-HLOS owned alarm
rtc: pm8xxx: mitigate flash wear
rtc: pm8xxx: add support for uefi offset
dt-bindings: rtc: qcom-pm8xxx: document qcom,no-alarm flag
rtc: rv3032: drop WADA
rtc: rv3032: fix EERD location
rtc: pm8xxx: switch to devm_device_init_wakeup
rtc: pm8xxx: fix possible race condition
rtc: mpfs: switch to devm_device_init_wakeup
...
net: airoha: Validate egress gdm port in airoha_ppe_foe_entry_prepare()
Dev pointer in airoha_ppe_foe_entry_prepare routine is not strictly
a device allocated by airoha_eth driver since it is an egress device
and the flowtable can contain even wlan, pppoe or vlan devices. E.g:
In this case airoha_get_dsa_port() will just return the original device
pointer and we can't assume netdev priv pointer points to an
airoha_gdm_port struct.
Fix the issue validating egress gdm port in airoha_ppe_foe_entry_prepare
routine before accessing net_device priv pointer.
net: dsa: mv88e6xxx: propperly shutdown PPU re-enable timer on destroy
The mv88e6xxx has an internal PPU that polls PHY state. If we want to
access the internal PHYs, we need to disable the PPU first. Because
that is a slow operation, a 10ms timer is used to re-enable it,
canceled with every access, so bulk operations effectively only
disable it once and re-enable it some 10ms after the last access.
If a PHY is accessed and then the mv88e6xxx module is removed before
the 10ms are up, the PPU re-enable ends up accessing a dangling pointer.
This especially affects probing during bootup. The MDIO bus and PHY
registration may succeed, but registration with the DSA framework
may fail later on (e.g. because the CPU port depends on another,
very slow device that isn't done probing yet, returning -EPROBE_DEFER).
In this case, probe() fails, but the MDIO subsystem may already have
accessed the MIDO bus or PHYs, arming the timer.
This is fixed as follows:
- If probe fails after mv88e6xxx_phy_init(), make sure we also call
mv88e6xxx_phy_destroy() before returning
- In mv88e6xxx_remove(), make sure we do the teardown in the correct
order, calling mv88e6xxx_phy_destroy() after unregistering the
switch device.
- In mv88e6xxx_phy_destroy(), destroy both the timer and the work item
that the timer might schedule, synchronously waiting in case one of
the callbacks already fired and destroying the timer first, before
waiting for the work item.
- Access to the PPU is guarded by a mutex, the worker acquires it
with a mutex_trylock(), not proceeding with the expensive shutdown
if that fails. We grab the mutex in mv88e6xxx_phy_destroy() to make
sure the slow PPU shutdown is already done or won't even enter, when
we wait for the work item.
Fixes: 2e5f032095ff ("dsa: add support for the Marvell 88E6131 switch chip") Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Link: https://patch.msgid.link/20250401135705.92760-1-david.oberhollenzer@sigma-star.at Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ipv6: fix omitted netlink attributes when using RTEXT_FILTER_SKIP_STATS
Using RTEXT_FILTER_SKIP_STATS is incorrectly skipping non-stats IPv6
netlink attributes on link dump. This causes issues on userspace tools,
e.g iproute2 is not rendering address generation mode as it should due
to missing netlink attribute.
Move the filling of IFLA_INET6_STATS and IFLA_INET6_ICMP6STATS to a
helper function guarded by a flag check to avoid hitting the same
situation in the future.
Fixes: d5566fd72ec1 ("rtnetlink: RTEXT_FILTER_SKIP_STATS support to avoid dumping inet/inet6 stats") Signed-off-by: Fernando Fernandez Mancera <ffmancera@riseup.net> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250402121751.3108-1-ffmancera@riseup.net Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When queue is being reset, callbacks of mgmt_ops are called by
netdev_nl_bind_rx_doit().
The netdev_nl_bind_rx_doit() first acquires netdev_lock() and then calls
callbacks.
So, mgmt_ops callbacks should not acquire netdev_lock() internaly.
The bnxt_queue_{start | stop}() calls napi_{enable | disable}() but they
internally acquire netdev_lock().
So, deadlock occurs.
To avoid deadlock, napi_{enable | disable}_locked() should be used
instead.
Signed-off-by: Taehee Yoo <ap420073@gmail.com> Acked-by: Stanislav Fomichev <sdf@fomichev.me> Reviewed-by: Michael Chan <michael.chan@broadcom.com> Fixes: cae03e5bdd9e ("net: hold netdev instance lock during queue operations") Link: https://patch.msgid.link/20250402133123.840173-1-ap420073@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net/selftests: Add loopback link local route for self-connect
self-connect-ipv6 got slightly flaky on netdev:
> # timeout set to 120
> # selftests: net/tcp_ao: self-connect_ipv6
> # 1..5
> # # 708[lib/setup.c:250] rand seed 1742872572
> # TAP version 13
> # # 708[lib/proc.c:213] Snmp6 Ip6OutNoRoutes: 0 => 1
> # not ok 1 # error 708[self-connect.c:70] failed to connect()
> # ok 2 No unexpected trace events during the test run
> # # Planned tests != run tests (5 != 2)
> # # Totals: pass:1 fail:0 xfail:0 xpass:0 skip:0 error:1
> ok 1 selftests: net/tcp_ao: self-connect_ipv6
I can not reproduce it on my machines, but judging by "Ip6OutNoRoutes"
there is no route to the local_addr (::1).
Looking at the kernel code, I see that kernel does add link-local
address automatically in init_loopback(), but that is called from
ipv6 notifier block. So, in turn the userspace that brought up
the loopback interface may see rtnetlink ACK earlier than
addrconf_notify() does it's job (at least, on a slow VM such as netdev).
Probably, for ipv4 it's the same, judging by inetdev_event().
The fix is quite simple: set the link-local route straight after
bringing the loopback interface. That will make it synchronous.
Edward Cree [Tue, 1 Apr 2025 22:54:39 +0000 (23:54 +0100)]
sfc: fix NULL dereferences in ef100_process_design_param()
Since cited commit, ef100_probe_main() and hence also
ef100_check_design_params() run before efx->net_dev is created;
consequently, we cannot netif_set_tso_max_size() or _segs() at this
point.
Move those netif calls to ef100_probe_netdev(), and also replace
netif_err within the design params code with pci_err.
Reported-by: Kyungwook Boo <bookyungwook@gmail.com> Fixes: 98ff4c7c8ac7 ("sfc: Separate netdev probe/remove from PCI probe/remove") Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Link: https://patch.msgid.link/20250401225439.2401047-1-edward.cree@amd.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
gve: handle overflow when reporting TX consumed descriptors
When the tx tail is less than the head (in cases of wraparound), the TX
consumed descriptor statistic in DQ will be reported as
UINT32_MAX - head + tail, which is incorrect. Mask the difference of
head and tail according to the ring size when reporting the statistic.
Cc: stable@vger.kernel.org Fixes: 2c9198356d56 ("gve: Add consumed counts to ethtool stats") Signed-off-by: Joshua Washington <joshwash@google.com> Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250402001037.2717315-1-hramamurthy@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux
Pull ARM and clkdev updates from Russell King:
- Simplify ARM_MMU_KEEP usage
- Add Rust support for ARM architecture version 7
- Align IPIs reported in /proc/interrupts
- require linker to support KEEP within OVERLAY
- add KEEP() for ARM vectors
- add __printf() attribute for clkdev functions
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux:
ARM: 9445/1: clkdev: Mark some functions with __printf() attribute
ARM: 9444/1: add KEEP() keyword to ARM_VECTORS
ARM: 9443/1: Require linker to support KEEP within OVERLAY for DCE
ARM: 9442/1: smp: Fix IPI alignment in /proc/interrupts
ARM: 9441/1: rust: Enable Rust support for ARMv7
ARM: 9439/1: arm32: simplify ARM_MMU_KEEP usage
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Catalin Marinas:
- Fix max_pfn calculation when hotplugging memory so that it never
decreases
- Fix dereference of unused source register in the MOPS SET operation
fault handling
- Fix NULL calling in do_compat_alignment_fixup() when the 32-bit user
space does an unaligned LDREX/STREX
- Add the HiSilicon HIP09 processor to the Spectre-BHB affected CPUs
- Drop unused code pud accessors (special/mkspecial)
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: Don't call NULL in do_compat_alignment_fixup()
arm64: Add support for HIP09 Spectre-BHB mitigation
arm64: mm: Drop dead code for pud special bit handling
arm64: mops: Do not dereference src reg for a set operation
arm64: mm: Correct the update of max_pfn
Merge tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Pull bpf fixes from Alexei Starovoitov:
- Fix BPF selftests expectations of assembler output and struct layout
(Song Liu and Yonghong Song)
- Fix XSK error code when queue is full (Wang Liang)
* tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
selftests/bpf: Fix verifier_private_stack test failure
selftests/bpf: Fix verifier_bpf_fastcall test
selftests/bpf: Fix tests after fields reorder in struct file
xsk: Fix __xsk_generic_xmit() error code when cq is full
Merge tag 'mm-nonmm-stable-2025-04-02-22-12' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull more non-MM updates from Andrew Morton:
"One bugfix and a couple of small late-arriving updates"
* tag 'mm-nonmm-stable-2025-04-02-22-12' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
lib: scatterlist: fix sg_split_phys to preserve original scatterlist offsets
lib/sort.c: add _nonatomic() variants with cond_resched()
mailmap: add an entry for Nicolas Schier
Merge tag 'mm-stable-2025-04-02-22-07' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull more MM updates from Andrew Morton:
- The series "mm: fixes for fallouts from mem_init() cleanup" from Mike
Rapoport fixes a couple of issues with the just-merged "arch, mm:
reduce code duplication in mem_init()" series
- The series "MAINTAINERS: add my isub-entries to MM part." from Mike
Rapoport does some maintenance on MAINTAINERS
- The series "remove tlb_remove_page_ptdesc()" from Qi Zheng does some
cleanup work to the page mapping code
- The series "mseal system mappings" from Jeff Xu permits sealing of
"system mappings", such as vdso, vvar, vvar_vclock, vectors (arm
compat-mode), sigpage (arm compat-mode)
- Plus the usual shower of singleton patches
* tag 'mm-stable-2025-04-02-22-07' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (31 commits)
mseal sysmap: add arch-support txt
mseal sysmap: enable s390
selftest: test system mappings are sealed
mseal sysmap: update mseal.rst
mseal sysmap: uprobe mapping
mseal sysmap: enable arm64
mseal sysmap: enable x86-64
mseal sysmap: generic vdso vvar mapping
selftests: x86: test_mremap_vdso: skip if vdso is msealed
mseal sysmap: kernel config and header change
mm: pgtable: remove tlb_remove_page_ptdesc()
x86: pgtable: convert to use tlb_remove_ptdesc()
riscv: pgtable: unconditionally use tlb_remove_ptdesc()
mm: pgtable: convert some architectures to use tlb_remove_ptdesc()
mm: pgtable: change pt parameter of tlb_remove_ptdesc() to struct ptdesc*
mm: pgtable: make generic tlb_remove_table() use struct ptdesc
microblaze/mm: put mm_cmdline_setup() in .init.text section
mm/memory_hotplug: fix call folio_test_large with tail page in do_migrate_range
MAINTAINERS: mm: add entry for secretmem
MAINTAINERS: mm: add entry for numa memblocks and numa emulation
...
Merge tag 'mm-hotfixes-stable-2025-04-02-21-57' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull MM hotfixes from Andrew Morton:
"Five hotfixes. Three are cc:stable and the remainder address post-6.14
issues or aren't considered necessary for -stable kernels.
All patches are for MM"
* tag 'mm-hotfixes-stable-2025-04-02-21-57' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
mm: zswap: fix crypto_free_acomp() deadlock in zswap_cpu_comp_dead()
mm/hugetlb: move hugetlb_sysctl_init() to the __init section
mm: page_isolation: avoid calling folio_hstate() without hugetlb_lock
mm/hugetlb_vmemmap: fix memory loads ordering
mm/userfaultfd: fix release hang over concurrent GUP
Merge tag 'sched_ext-for-6.15-rc0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- Calling scx_bpf_create_dsq() with the same ID would succeed creating
duplicate DSQs. Fix it to return -EEXIST.
- scx_select_cpu_dfl() fixes and cleanups.
- Synchronize tool/sched_ext with external scheduler repo. While this
isn't a fix. There's no risk to the kernel and it's better if they
stay synced closer.
* tag 'sched_ext-for-6.15-rc0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
tools/sched_ext: Sync with scx repo
sched_ext: initialize built-in idle state before ops.init()
sched_ext: create_dsq: Return -EEXIST on duplicate request
sched_ext: Remove a meaningless conditional goto in scx_select_cpu_dfl()
sched_ext: idle: Fix return code of scx_select_cpu_dfl()
Merge tag 'trace-v6.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing fixes from Steven Rostedt:
- Fix build error when CONFIG_PROBE_EVENTS_BTF_ARGS is not enabled
The tracing of arguments in the function tracer depends on some
functions that are only defined when PROBE_EVENTS_BTF_ARGS is
enabled. In fact, PROBE_EVENTS_BTF_ARGS also depends on all the same
configs as the function argument tracing requires. Just have the
function argument tracing depend on PROBE_EVENTS_BTF_ARGS.
- Free module_delta for persistent ring buffer instance
When an instance holds the persistent ring buffer, it allocates a
helper array to hold the deltas between where modules are loaded on
the last boot and the current boot. This array needs to be freed when
the instance is freed.
- Add cond_resched() to loop in ftrace_graph_set_hash()
The hash functions in ftrace loop over every function that can be
enabled by ftrace. This can be 50,000 functions or more. This loop is
known to trigger soft lockup warnings and requires a cond_resched().
The loop in ftrace_graph_set_hash() was missing it.
- Fix the event format verifier to include "%*p.." arguments
To prevent events from dereferencing stale pointers that can happen
if a trace event uses a dereferece pointer to something that was not
copied into the ring buffer and can be freed by the time the trace is
read, a verifier is called. At boot or module load, the verifier
scans the print format string for pointers that can be dereferenced
and it checks the arguments to make sure they do not contain
something that can be freed. The "%*p" was not handled, which would
add another argument and cause the verifier to not only not verify
this pointer, but it will look at the wrong argument for every
pointer after that.
- Fix mcount sorttable building for different endian type target
When modifying the ELF file to sort the mcount_loc table in the
sorttable.c code, the endianess of the file and the host is used to
determine if the bytes need to be swapped when calculations are done.
A change was made to the sorting of the mcount_loc that read the
values from the ELF file into an array and the swap happened on the
filling of the array. But one of the calculations of the array still
did the swap when it did not need to. This caused building on a
little endian machine for a big endian target to not find the mcount
function in the 'nm' table and it zeroed it out, causing there to be
no functions available to trace.
- Add goto out_unlock jump to rv_register_monitor() on error path
One of the error paths in rv_register_monitor() just returned the
error when it should have jumped to the out_unlock label to release
the mutex.
* tag 'trace-v6.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
rv: Fix missing unlock on double nested monitors return path
scripts/sorttable: Fix endianness handling in build-time mcount sort
tracing: Verify event formats that have "%*p.."
ftrace: Add cond_resched() to ftrace_graph_set_hash()
tracing: Free module_delta on freeing of persistent ring buffer
ftrace: Have tracing function args depend on PROBE_EVENTS_BTF_ARGS
Kent Overstreet [Wed, 2 Apr 2025 16:23:25 +0000 (12:23 -0400)]
bcachefs: Fix "journal stuck" during recovery
If we crash when the journal pin fifo is completely full - i.e. we're
at the maximum number of dirty journal entries - that may put us in a
sticky situation in recovery, as journal replay will need to be able to
open new journal entries in order to get going.
bch2_fs_journal_start() already had provisions for resizing the journal
pin fifo if needed, but it needs a fudge factor to ensure there's room
for journal replay.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Codepaths that create entries in the snapshots btree currently call
bch2_mark_snapshot(), which updates the in-memory snapshot table, before
transaction commit.
This is because bch2_mark_snapshot() is an atomic trigger, run with
btree write locks held, and isn't allowed to fail - but it might need to
reallocate the table, hence we call it early when we're still allowed to
fail.
This is generally harmless - if we fail, we'll have left an entry in the
snapshots table around, but nothing will reference it and it'll get
overwritten if reused by another transaction.
But check_snapshot_exists(), which reconstructs snapshots when the
snapshots btree has been corrupted or lost, was erronously rechecking if
the snapshot exists inside the transaction commit loop - so on
transaction restart (in this case mem_realloced), the second iteration
would return without repairing.
This code needs some cleanup: splitting out a "maybe realloc snapshots
table" helper would have avoided this, that will be in the next patch.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
bcachefs: use nonblocking variant of print_string_as_lines in error path
The inconsistency error path calls print_string_as_lines, which calls
console_lock, which is a potentially-sleeping function and so can't be
called in an atomic context.
Replace calls to it with the nonblocking variant which is safe to call.
Signed-off-by: Bharadwaj Raju <bharadwaj.raju777@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Kent Overstreet [Tue, 1 Apr 2025 17:01:29 +0000 (13:01 -0400)]
bcachefs: Fix scheduling while atomic from logging changes
Two fixes from the recent logging changes:
bch2_inconsistent(), bch2_fs_inconsistent() be called from interrupt
context, or with rcu_read_lock() held.
The one syzbot found is in
bch2_bkey_pick_read_device
bch2_dev_rcu
bch2_fs_inconsistent
We're starting to switch to lift the printbufs up to higher levels so we
can emit better log messages and print them all in one go (avoid
garbling), so that conversion will help with spotting these in the
future; when we declare a printbuf it must be flagged if we're in an
atomic context.
Ming Lei [Thu, 3 Apr 2025 10:54:02 +0000 (18:54 +0800)]
block: don't grab elevator lock during queue initialization
->elevator_lock depends on queue freeze lock, see block/blk-sysfs.c.
queue freeze lock depends on fs_reclaim.
So don't grab elevator lock during queue initialization which needs to
call kmalloc(GFP_KERNEL), and we can cut the dependency between
->elevator_lock and fs_reclaim, then the lockdep warning can be killed.
This way is safe because elevator setting isn't ready to run during
queue initialization.
There isn't such issue in __blk_mq_update_nr_hw_queues() because
memalloc_noio_save() is called before acquiring elevator lock.
Pavel Begunkov [Thu, 3 Apr 2025 11:29:30 +0000 (12:29 +0100)]
io_uring: always do atomic put from iowq
io_uring always switches requests to atomic refcounting for iowq
execution before there is any parallilism by setting REQ_F_REFCOUNT,
and the flag is not cleared until the request completes. That should be
fine as long as the compiler doesn't make up a non existing value for
the flags, however KCSAN still complains when the request owner changes
oter flag bits:
BUG: KCSAN: data-race in io_req_task_cancel / io_wq_free_work
...
read to 0xffff888117207448 of 8 bytes by task 3871 on cpu 0:
req_ref_put_and_test io_uring/refs.h:22 [inline]
Skip REQ_F_REFCOUNT checks for iowq, we know it's set.
Lin Ma [Wed, 2 Apr 2025 17:00:26 +0000 (01:00 +0800)]
netfilter: nft_tunnel: fix geneve_opt type confusion addition
When handling multiple NFTA_TUNNEL_KEY_OPTS_GENEVE attributes, the
parsing logic should place every geneve_opt structure one by one
compactly. Hence, when deciding the next geneve_opt position, the
pointer addition should be in units of char *.
However, the current implementation erroneously does type conversion
before the addition, which will lead to heap out-of-bounds write.
Fix this bug with correct pointer addition and conversion in parse
and dump code.
Fixes: 925d844696d9 ("netfilter: nft_tunnel: add support for geneve opts") Signed-off-by: Lin Ma <linma@zju.edu.cn> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Antoine Tenart [Wed, 26 Mar 2025 17:36:32 +0000 (18:36 +0100)]
net: decrease cached dst counters in dst_release
Upstream fix ac888d58869b ("net: do not delay dst_entries_add() in
dst_release()") moved decrementing the dst count from dst_destroy to
dst_release to avoid accessing already freed data in case of netns
dismantle. However in case CONFIG_DST_CACHE is enabled and OvS+tunnels
are used, this fix is incomplete as the same issue will be seen for
cached dsts:
Merge tag 'firewire-updates-6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394
Pull firewire update from Takashi Sakamoto:
"A single commit to use the common helper function for on-stack
trailing array to enqueue any isochronous packet by the requests
from userspace applications"
* tag 'firewire-updates-6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:
firewire: core: avoid -Wflex-array-member-not-at-end warning
As a result, verifier_bpf_fastcall tests should now expect a negative
value for pcpu_hot, IOW, the disassemble should show "r=" instead of
"w=".
Fix this in the test.
Note that, a later change created a new variable "cpu_number" for
bpf_get_smp_processor_id() [2]. The inlining logic is updated properly
as part of this change, so there is no need to fix anything on the
kernel side.
[1] commit 9d7de2aa8b41 ("x86/percpu/64: Use relative percpu offsets")
[2] commit 01c7bc5198e9 ("x86/smp: Move cpu number to percpu hot section") Reported-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20250328193124.808784-1-song@kernel.org Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Song Liu [Thu, 27 Mar 2025 18:55:28 +0000 (11:55 -0700)]
selftests/bpf: Fix tests after fields reorder in struct file
The change in struct file [1] moved f_ref to the 3rd cache line.
It made *(u64 *)file dereference invalid from the verifier point of view,
because btf_struct_walk() walks into f_lock field, which is 4-byte long.
Fix the selftests to deference the file pointer as a 4-byte access.
[1] commit e249056c91a2 ("fs: place f_ref to 3rd cache line in struct file to resolve false sharing") Reported-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20250327185528.1740787-1-song@kernel.org Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Wang Liang [Thu, 27 Feb 2025 08:10:52 +0000 (16:10 +0800)]
xsk: Fix __xsk_generic_xmit() error code when cq is full
When the cq reservation is failed, the error code is not set which is
initialized to zero in __xsk_generic_xmit(). That means the packet is not
send successfully but sendto() return ok.
Considering the impact on uapi, return -EAGAIN is a good idea. The cq is
full usually because it is not released in time, try to send msg again is
appropriate.
The bug was at the very early implementation of xsk, so the Fixes tag
targets the commit that introduced the changes in
xsk_cq_reserve_addr_locked where this fix depends on.
Fixes: e6c4047f5122 ("xsk: Use xsk_buff_pool directly for cq functions") Suggested-by: Magnus Karlsson <magnus.karlsson@gmail.com> Signed-off-by: Wang Liang <wangliang74@huawei.com> Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org> Acked-by: Stanislav Fomichev <sdf@fomichev.me> Link: https://patch.msgid.link/20250227081052.4096337-1-wangliang74@huawei.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
- dm-integrity: set ti->error on memory allocation failure
- dm-bufio: remove unused return value
- dm-verity: do forward error correction on metadata I/O errors
- dm: fix unconditional IO throttle caused by REQ_PREFLUSH
- dm cache: prevent BUG_ON by blocking retries on failed device resumes
- dm cache: support shrinking the origin device
- dm: restrict dm device size to 2^63-512 bytes
- dm-delay: support zoned devices
- dm-verity: support block number limits for different ioprio classes
- dm-integrity: fix non-constant-time tag verification (security bug)
- dm-verity, dm-ebs: fix prefetch-vs-suspend race
* tag 'for-6.15/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (27 commits)
dm-ebs: fix prefetch-vs-suspend race
dm-verity: fix prefetch-vs-suspend race
dm-integrity: fix non-constant-time tag verification
dm-verity: support block number limits for different ioprio classes
dm-delay: support zoned devices
dm: restrict dm device size to 2^63-512 bytes
dm cache: support shrinking the origin device
dm cache: prevent BUG_ON by blocking retries on failed device resumes
dm vdo indexer: reorder uds_request to reduce padding
dm: fix unconditional IO throttle caused by REQ_PREFLUSH
dm vdo: rework processing of loaded refcount byte arrays
dm vdo: remove remaining ring references
dm-verity: do forward error correction on metadata I/O errors
dm-bufio: remove unused return value
dm-integrity: set ti->error on memory allocation failure
dm: Enable inline crypto passthrough for striped target
dm vdo slab-depot: read refcount blocks in large chunks at load time
dm vdo vio-pool: allow variable-sized metadata vios
dm vdo vio-pool: support pools with multiple data blocks per vio
dm vdo vio-pool: add a pool pointer to pooled_vio
...
projects / thirdparty / linux.git / log
