NETC IP has only one external master MDIO interface (eMDIO) for managing
the external PHYs. ENETC can use the interfaces provided by the EMDIO
function or its port MDIO to access and manage its external PHY. Both
the EMDIO function and the port MDIO are all virtual ports of the eMDIO.
The difference is that the EMDIO function is a 'global port', it can
access all the PHYs on the eMDIO, but port MDIO can only access its own
PHY. To ensure that ENETC can only access its own PHY through port MDIO,
LaBCR[MDIO_PHYAD_PRTAD] needs to be set, which represents the address of
the external PHY connected to ENETC. If the accessed PHY address is not
consistent with LaBCR[MDIO_PHYAD_PRTAD], then the MDIO access initiated
by port MDIO will be invalid.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Link: https://patch.msgid.link/20251119102557.1041881-3-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The ENETC supports managing its own external PHY through its port MDIO
functionality. To use this function, the PHY address needs be set in the
corresponding LaBCR register in the Integrated Endpoint Register Block
(IERB), which is used for pre-boot initialization of NETC PCIe functions.
The port MDIO can only work properly when the PHY address accessed by the
port MDIO matches the corresponding LaBCR[MDIO_PHYAD_PRTAD] value.
Because the ENETC driver only registers the MDIO bus (port MDIO bus) when
it detects an MDIO child node in its node, similarly, the netc-blk-ctrl
driver only resolves the PHY address and sets it in the corresponding
LaBCR when it detects an MDIO child node in the ENETC node.
Co-developed-by: Aziz Sellami <aziz.sellami@nxp.com>
Signed-off-by: Aziz Sellami <aziz.sellami@nxp.com>
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Link: https://patch.msgid.link/20251119102557.1041881-2-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Vladimir Oltean says:
====================
Improvements over DSA conduit ethtool ops
DSA interceps 'ethtool -S eth0', where eth0 is the host port of the
switch (called 'conduit'). It does this because otherwise there is no
way to report port counters for the CPU port, which is a MAC like any
other of that switch, except Linux exposes no net_device for it, thus no
ethtool hook.
Having understood all downsides of this debugging interface, when we
need it we needed, so the proposed changes here are to make it more
useful by dumping more counters in it: not just the switch CPU port,
but all other switch ports in the tree which lack a net_device. Not
reinventing any wheel, just putting more output in an existing command.
That is patch 3/3. The other 2 are cleanup.
====================
Link: https://patch.msgid.link/20251122112311.138784-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently there is no way to see packet counters on cascade ports, and
no clarity on how the API for that would look like.
Because it's something that is currently needed, just extend the hack
where ethtool -S on the conduit interface dumps CPU port counters, and
also use it to dump counters of cascade ports.
Note that the "pXX_" naming convention changes to "sXX_pYY", to
distinguish between ports having the same index but belonging to
different switches. This has a slight chance of causing regressions to
existing tooling:
- grepping for "p04_counter_name" still works, but might return more
than one string now
- grepping for " p04_counter_name" no longer works
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20251122112311.138784-4-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Suppress some checkpatch 'CHECK' messages about u8 being preferable over
uint8_t, etc. No functional change.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20251122112311.138784-3-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In theory this would have been seen by now, but it seems that all
drivers used as DSA conduit interfaces thus far have had ethtool_ops
set, and it's hard to even find modern Ethernet drivers (and not VF
ones) which don't use ethtool.
Here is the unfiltered list of drivers which register any sort of
net_device but don't set its ethtool_ops pointer. I don't think any of
them 'risks' being used as a DSA conduit, maybe except for moxart,
rnpbge and icssm, I'm not sure.
- drivers/net/can/dev/dev.c
- drivers/net/wwan/qcom_bam_dmux.c
- drivers/net/wwan/t7xx/t7xx_netdev.c
- drivers/net/arcnet/arcnet.c
- drivers/net/hamradio/
- drivers/net/slip/slip.c
- drivers/net/ethernet/ezchip/nps_enet.c
- drivers/net/ethernet/moxa/moxart_ether.c
- drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c
- drivers/net/ethernet/wangxun/ngbevf/ngbevf_main.c
- drivers/net/ethernet/huawei/hinic3/hinic3_main.c
- drivers/net/ethernet/i825xx/
- drivers/net/ethernet/ti/icssm/icssm_prueth.c
- drivers/net/ethernet/seeq/
- drivers/net/ethernet/litex/litex_liteeth.c
- drivers/net/ethernet/sunplus/spl2sw_driver.c
- drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c
- drivers/net/ipa/
- drivers/net/wireless/microchip/wilc1000/
- drivers/net/wireless/mediatek/mt76/dma.c
- drivers/net/wireless/ath/ath12k/
- drivers/net/wireless/ath/ath11k/
- drivers/net/wireless/ath/ath6kl/
- drivers/net/wireless/ath/ath10k/
- drivers/net/wireless/intel/iwlwifi/pcie/gen1_2/trans.c
- drivers/net/wireless/virtual/mac80211_hwsim.c
- drivers/net/wireless/quantenna/qtnfmac/pcie/pcie.c
- drivers/net/wireless/realtek/rtw89/core.c
- drivers/net/wireless/realtek/rtw88/pci.c
- drivers/net/caif/
- drivers/net/plip/
- drivers/net/wan/
- drivers/net/mctp/
- drivers/net/ppp/
- drivers/net/thunderbolt/
Nonetheless, it's good for the framework not to make such assumptions,
and not panic when coming across such kind of host device in the future.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20251122112311.138784-2-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Attempting to add a port device that is already up will expectedly fail,
but not before modifying the team device header_ops.
In the case of the syzbot reproducer the gre0 device is
already in state UP when it attempts to add it as a
port device of team0, this fails but before that
header_ops->create of team0 is changed from eth_header to ipgre_header
in the call to team_dev_type_check_change.
Later when we end up in ipgre_header() struct ip_tunnel* points to nonsense
as the private data of the device still holds a struct team.
Example sequence of iproute2 commands to reproduce the hang/BUG():
ip link add dev team0 type team
ip link add dev gre0 type gre
ip link set dev gre0 up
ip link set dev gre0 master team0
ip link set dev team0 up
ping -I team0 1.1.1.1
Move team_dev_type_check_change down where all other checks have passed
as it changes the dev type with no way to restore it in case
one of the checks that follow it fail.
Also make sure to preserve the origial mtu assignment:
- If port_dev is not the same type as dev, dev takes mtu from port_dev
- If port_dev is the same type as dev, port_dev takes mtu from dev
This is done by adding a conditional before the call to dev_set_mtu
to prevent it from assigning port_dev->mtu = dev->mtu and instead
letting team_dev_type_check_change assign dev->mtu = port_dev->mtu.
The conditional is needed because the patch moves the call to
team_dev_type_check_change past dev_set_mtu.
Testing:
- team device driver in-tree selftests
- Add/remove various devices as slaves of team device
- syzbot
Reported-by: syzbot+a2a3b519de727b0f7903@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=a2a3b519de727b0f7903
Fixes: 1d76efe157 ("team: add support for non-ethernet devices")
Signed-off-by: Nikola Z. Ivanov <zlatistiv@gmail.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://patch.msgid.link/20251122002027.695151-1-zlatistiv@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
drivers/net/ethernet/chelsio/cxgb4/sched.h declares a sched_class
struct which has a type name clash with struct sched_class
in kernel/sched/sched.h (a type used in a field in task_struct).
When cxgb4 is a builtin we end up with both sched_class types,
and as a result of this we wind up with DWARF (and derived from
that BTF) with a duplicate incorrect task_struct representation.
When cxgb4 is built-in this type clash can cause kernel builds to
fail as resolve_btfids will fail when confused which task_struct
to use. See [1] for more details.
As such, renaming sched_class to ch_sched_class (in line with
other structs like ch_sched_flowc) makes sense.
[1] https://lore.kernel.org/bpf/2412725b-916c-47bd-91c3-c2d57e3e6c7b@acm.org/
Reported-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Potnuri Bharat Teja <bharat@chelsio.com>
Link: https://patch.msgid.link/20251121181231.64337-1-alan.maguire@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This adds support for chip RTL9151A. Its XID is 0x68b. It is bascially
basd on the one with XID 0x688, but with different firmware file.
Signed-off-by: Javen Xu <javen_xu@realsil.com.cn>
Reviewed-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://patch.msgid.link/20251121090104.3753-1-javen_xu@realsil.com.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The rate limiting validation condition currently checks the output
variable max_bw_value[i] instead of the input value
maxrate->tc_maxrate[i]. This causes the validation to compare an
uninitialized or stale value rather than the actual requested rate.
The condition should check the input rate to properly validate against
the upper limit:
} else if (maxrate->tc_maxrate[i] <= upper_limit_gbps) {
This aligns with the pattern used in the first branch, which correctly
checks maxrate->tc_maxrate[i] against upper_limit_mbps.
The current implementation can lead to unreliable validation behavior:
- For rates between 25.5 Gbps and 255 Gbps, if max_bw_value[i] is 0
from initialization, the GBPS path may be taken regardless of whether
the actual rate is within bounds
- When processing multiple TCs (i > 0), max_bw_value[i] contains the
value computed for the previous TC, affecting the validation logic
- The overflow check for rates exceeding 255 Gbps may not trigger
consistently depending on previous array values
This patch ensures the validation correctly examines the requested rate
value for proper bounds checking.
Fixes: 43b27d1bd8 ("net/mlx5e: Fix wraparound in rate limiting for values above 255 Gbps")
Signed-off-by: Danielle Costantino <dcostantino@meta.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Link: https://patch.msgid.link/20251124180043.2314428-1-dcostantino@meta.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When having a multiuser mount with domain= specified and using
cifscreds, cifs_set_cifscreds() will end up setting @ctx->domainname,
so it needs to be freed before leaving cifs_construct_tcon().
This fixes the following memory leak reported by kmemleak:
mount.cifs //srv/share /mnt -o domain=ZELDA,multiuser,...
su - testuser
cifscreds add -d ZELDA -u testuser
...
ls /mnt/1
...
umount /mnt
echo scan > /sys/kernel/debug/kmemleak
cat /sys/kernel/debug/kmemleak
unreferenced object 0xffff8881203c3f08 (size 8):
comm "ls", pid 5060, jiffies 4307222943
hex dump (first 8 bytes):
5a 45 4c 44 41 00 cc cc ZELDA...
backtrace (crc d109a8cf):
__kmalloc_node_track_caller_noprof+0x572/0x710
kstrdup+0x3a/0x70
cifs_sb_tlink+0x1209/0x1770 [cifs]
cifs_get_fattr+0xe1/0xf50 [cifs]
cifs_get_inode_info+0xb5/0x240 [cifs]
cifs_revalidate_dentry_attr+0x2d1/0x470 [cifs]
cifs_getattr+0x28e/0x450 [cifs]
vfs_getattr_nosec+0x126/0x180
vfs_statx+0xf6/0x220
do_statx+0xab/0x110
__x64_sys_statx+0xd5/0x130
do_syscall_64+0xbb/0x380
entry_SYSCALL_64_after_hwframe+0x77/0x7f
Fixes: f2aee329a6 ("cifs: set domainName when a domain-key is used in multiuser")
Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Reviewed-by: David Howells <dhowells@redhat.com>
Cc: Jay Shin <jaeshin@redhat.com>
Cc: stable@vger.kernel.org
Cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
Kumar Kartikeya Dwivedi says:
====================
General enhancements to rqspinlock stress test
Three enchancements, details in commit messages.
First, the CPU requirements are 2 for AA, 3 for ABBA, and 4 for ABBCCA,
hence relax the check during module initialization. Second, add a
per-CPU histogram to capture lock acquisition times to record which
buckets these acquisitions fall into for the normal task context and NMI
context. Anything below 10ms is not printed in detail, but above that
displays the full breakdown for each context. Finally, make the delay of
the NMI and task contexts configurable, set to 10 and 20 ms respectively
by default.
====================
Link: https://patch.msgid.link/20251125020749.2421610-1-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Allow users to configure the critical section delay for both task/normal
and NMI contexts, and set to 20ms and 10ms as before by default.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20251125020749.2421610-4-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add statistics per-CPU broken down by context and various timing windows
for the time taken to acquire an rqspinlock. Cases where all
acquisitions fit into the 10ms window are skipped from printing,
otherwise the full breakdown is displayed when printing the summary.
This allows capturing precisely the number of times outlier attempts
happened for a given lock in a given context.
A critical detail is that time is captured regardless of success or
failure, which is important to capture events for failed but long
waiting timeout attempts.
Output:
[ 64.279459] rqspinlock acquisition latency histogram (ms):
[ 64.279472] cpu1: total 528426 (normal 526559, nmi 1867)
[ 64.279477] 0-1ms: total 524697 (normal 524697, nmi 0)
[ 64.279480] 2-2ms: total 3652 (normal 1811, nmi 1841)
[ 64.279482] 3-3ms: total 66 (normal 47, nmi 19)
[ 64.279485] 4-4ms: total 2 (normal 1, nmi 1)
[ 64.279487] 5-5ms: total 1 (normal 1, nmi 0)
[ 64.279489] 6-6ms: total 1 (normal 0, nmi 1)
[ 64.279490] 101-150ms: total 1 (normal 0, nmi 1)
[ 64.279492] >= 251ms: total 6 (normal 2, nmi 4)
...
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20251125020749.2421610-3-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Only require 2 CPUs for AA, 3 for ABBA, 4 for ABBCCA, which is
calculated nicely by adding to the mode enum. Enables running single CPU
AA tests.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20251125020749.2421610-2-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
It is to unify map flags checking for lookup_elem, update_elem,
lookup_batch and update_batch APIs.
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
Link: https://lore.kernel.org/r/20251125145857.98134-2-leon.hwang@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Currently, HFS declares on-disk layout's metadata structures
in fs/hfs/hfs.h and HFS+ declares it in fs/hfsplus/hfsplus_raw.h.
However, HFS and HFS+ on-disk layouts have some similarity and
overlapping in declarations. As a result, fs/hfs/hfs.h and
fs/hfsplus/hfsplus_raw.h contain multiple duplicated declarations.
Moreover, both HFS and HFS+ drivers contain completely similar
implemented functionality in multiple places.
This patch is moving the on-disk layout declarations from
fs/hfs/hfs.h and fs/hfsplus/hfsplus_raw.h into
include/linux/hfs_common.h with the goal to exclude
the duplication in declarations. Also, this patch prepares
the basis for creating a hfslib that can aggregate common
functionality without necessity to duplicate the same code
in HFS and HFS+ drivers.
Signed-off-by: Viacheslav Dubeyko <slava@dubeyko.com>
cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
cc: Yangtao Li <frank.li@vivo.com>
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Viacheslav Dubeyko <slava@dubeyko.com>
The bench test "trig-kernel-count" can be used as a baseline comparison
for fentry and other benchmarks, and the calling to bpf_get_numa_node_id()
should be considered as composition of the baseline. So, let's call it in
trigger_count(). Meanwhile, rename trigger_count() to
trigger_kernel_count() to make it easier understand.
Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20251116014242.151110-1-dongml2@chinatelecom.cn
Specify value size limit for BPF_MAP_TYPE_PERCPU_ARRAY which
is PCPU_MIN_UNIT_SIZE (32 kb). In percpu allocator (mm: percpu),
any request with a size greater than PCPU_MIN_UNIT_SIZE is rejected.
Signed-off-by: Alex Tran <alex.t.tran@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20251115063531.2302903-1-alex.t.tran@gmail.com
Orange Pi RV is a SBC based on the StarFive VisionFive 2 board.
Orange Pi RV features:
- StarFive JH7110 SoC
- GbE port connected to JH7110 GMAC0 via YT8531 PHY
- 4x USB ports via VL805 PCIe USB controller connected to JH7110 pcie0
- M.2 M-key slot connected to JH7110 pcie1
- HDMI video output
- 3.5mm audio output
- Ampak AP6256 SDIO Wi-Fi/Bluetooth module on mmc0
- microSD slot on mmc1
- SPI NOR flash memory
- 24c02 EEPROM (read only by default)
Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
Signed-off-by: E Shattow <e@freeshell.de>
[conor: amend comment to say what's missing]
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Add a common board dtsi for use by VisionFive 2 Lite and
VisionFive 2 Lite eMMC.
Acked-by: Emil Renner Berthing <emil.renner.berthing@canonical.com>
Tested-by: Matthias Brugger <mbrugger@suse.com>
Signed-off-by: Hal Feng <hal.feng@starfivetech.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Some node in this file are not used by the upcoming VisionFive 2 Lite
board. Move them to the board dts to prepare for adding the new
VisionFive 2 Lite device tree.
Tested-by: Matthias Brugger <mbrugger@suse.com>
Signed-off-by: Hal Feng <hal.feng@starfivetech.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Add device tree bindings for the StarFive JH7110S SoC
and the VisionFive 2 Lite board equipped with it.
JH7110S SoC is an industrial SoC which can run at -40~85 degrees centigrade
and up to 1.25GHz. Its CPU cores and peripherals are the same as
those of the JH7110 SoC.
VisionFive 2 Lite boards have MicroSD card version (default) and eMMC
version, which are called "VisionFive 2 Lite" and "VisionFive 2 Lite eMMC"
respectively.
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Tested-by: Matthias Brugger <mbrugger@suse.com>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Signed-off-by: Hal Feng <hal.feng@starfivetech.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
The SiFive and Canaan platforms are not being actively looked after at
this point, but fixes for them would be applied if/when the patches
appeared. Since they're now the only things in the RISC-V MISC SOC
SUPPORT, mark them as Odd Fixes. I don't believe this is a functional
change, it just represents what's actually happening - particularly
since the Canaan k230 never built up enough steam to get merged and the
new SiFive demo chips have been done in partnership with with other
companies, e.g. Eswin, and will reside in their directories instead.
Reviewed-by: Paul Walmsley <pjw@kernel.org>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
In fairness to my own employer, lumping it in as "misc" is not quite
accurate when they do pay me to look after the platform. Move the tree
link for it to its entry, rather than having the RISC-V MISC SOC SUPPORT
entry cover it.
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
I don't use the main riscv patchwork for anything to do with SoCs,
remove them from here to avoid confusion.
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
I apply the patches for StarFive devicetrees, add me to the entry along
with my tree location etc. This is not a functional change, as this info
was in the "RISC-V MISC" entry but I'd rather not have the duplication
of entries covering the StarFive directory.
Acked-by: Emil Renner Berthing <kernel@esmil.dk>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Although commit 0c9992315e ("ACPICA: Avoid walking the ACPI Namespace
if it is not there") fixed the situation when both start_node and
acpi_gbl_root_node are NULL, the Linux kernel mainline now still crashed
on Honor Magicbook 14 Pro [1].
That happens due to the access to the member of parent_node in
acpi_ns_get_next_node(). The NULL pointer dereference will always
happen, no matter whether or not the start_node is equal to
ACPI_ROOT_OBJECT, so move the check of start_node being NULL
out of the if block.
Unfortunately, all the attempts to contact Honor have failed, they
refused to provide any technical support for Linux.
The bad DSDT table's dump could be found on GitHub [2].
DMI: HONOR FMB-P/FMB-P-PCB, BIOS 1.13 05/08/2025
Link: 1c1b57b9eb
Link: https://gist.github.com/Cryolitia/a860ffc97437dcd2cd988371d5b73ed7 [1]
Link: https://github.com/denis-bb/honor-fmb-p-dsdt [2]
Signed-off-by: Cryolitia PukNgae <cryolitia.pukngae@linux.dev>
Reviewed-by: WangYuli <wangyl5933@chinaunicom.cn>
[ rjw: Subject adjustment, changelog edits ]
Link: https://patch.msgid.link/20251125-acpica-v1-1-99e63b1b25f8@linux.dev
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
When a VMA is split (e.g., by partial munmap or MAP_FIXED), the kernel
calls vm_ops->close on each portion. For trace buffer mappings, this
results in ring_buffer_unmap() being called multiple times while
ring_buffer_map() was only called once.
This causes ring_buffer_unmap() to return -ENODEV on subsequent calls
because user_mapped is already 0, triggering a WARN_ON.
Trace buffer mappings cannot support partial mappings because the ring
buffer structure requires the complete buffer including the meta page.
Fix this by adding a may_split callback that returns -EINVAL to prevent
VMA splits entirely.
Cc: stable@vger.kernel.org
Fixes: cf9f0f7c4c ("tracing: Allow user-space mapping of the ring-buffer")
Link: https://patch.msgid.link/20251119064019.25904-1-kartikey406@gmail.com
Closes: https://syzkaller.appspot.com/bug?extid=a72c325b042aae6403c7
Tested-by: syzbot+a72c325b042aae6403c7@syzkaller.appspotmail.com
Reported-by: syzbot+a72c325b042aae6403c7@syzkaller.appspotmail.com
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
When tick counts are large and multiplication by MSEC_PER_SEC is larger
than 64 bits, the conversion from clock ticks to milliseconds can go bad.
Use mul_u64_u32_div() instead.
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Suggested-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Fixes: 49cc215aad ("drm/xe: Add xe_gt_clock_interval_to_ms helper")
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patch.msgid.link/1562f1b62d5be3fbaee100f09107f3cc49e40dd1.1763408584.git.harish.chegondi@intel.com
(cherry picked from commit 96b93ac214)
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
xe_guc_ct_init_noalloc() allocates the CT workqueue and other helpers
before it tries to initialize ct->lock. If drmm_mutex_init() fails
we currently bail out without releasing those resources because the
guc_ct_fini() hasn’t been registered yet.
Since destroy_workqueue() in guc_ct_fini() may flush the workqueue, which
in turn can take the ct lock, the initialization sequence is restructured
to first initialize the ct->lock, then set up all CT state, and finally
register guc_ct_fini().
v2: guc_ct_fini() does take ct lock. (Matt)
v3: move primelockdep() together with drmm_mutex_init(). (Lucas)
Fixes: dd08ebf6c3 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patch.msgid.link/20251110184522.1581001-2-shuicheng.lin@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit 2e4ad5b066)
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
The xfstests' test-case generic/101 leaves HFS+ volume
in corrupted state:
sudo ./check generic/101
FSTYP -- hfsplus
PLATFORM -- Linux/x86_64 hfsplus-testing-0001 6.17.0-rc1+ #4 SMP PREEMPT_DYNAMIC Wed Oct 1 15:02:44 PDT 2025
MKFS_OPTIONS -- /dev/loop51
MOUNT_OPTIONS -- /dev/loop51 /mnt/scratch
generic/101 _check_generic_filesystem: filesystem on /dev/loop51 is inconsistent
(see XFSTESTS-2/xfstests-dev/results//generic/101.full for details)
Ran: generic/101
Failures: generic/101
Failed 1 of 1 tests
sudo fsck.hfsplus -d /dev/loop51
** /dev/loop51
Using cacheBlockSize=32K cacheTotalBlock=1024 cacheSize=32768K.
Executing fsck_hfs (version 540.1-Linux).
** Checking non-journaled HFS Plus Volume.
The volume name is untitled
** Checking extents overflow file.
** Checking catalog file.
** Checking multi-linked files.
** Checking catalog hierarchy.
** Checking extended attributes file.
** Checking volume bitmap.
** Checking volume information.
Invalid volume free block count
(It should be 2614350 instead of 2614382)
Verify Status: VIStat = 0x8000, ABTStat = 0x0000 EBTStat = 0x0000
CBTStat = 0x0000 CatStat = 0x00000000
** Repairing volume.
** Rechecking volume.
** Checking non-journaled HFS Plus Volume.
The volume name is untitled
** Checking extents overflow file.
** Checking catalog file.
** Checking multi-linked files.
** Checking catalog hierarchy.
** Checking extended attributes file.
** Checking volume bitmap.
** Checking volume information.
** The volume untitled was repaired successfully.
This test executes such steps: "Test that if we truncate a file
to a smaller size, then truncate it to its original size or
a larger size, then fsyncing it and a power failure happens,
the file will have the range [first_truncate_size, last_size[ with
all bytes having a value of 0x00 if we read it the next time
the filesystem is mounted.".
HFS+ keeps volume's free block count in the superblock.
However, hfsplus_file_fsync() doesn't store superblock's
content. As a result, superblock contains not correct
value of free blocks if a power failure happens.
This patch adds functionality of saving superblock's
content during hfsplus_file_fsync() call.
sudo ./check generic/101
FSTYP -- hfsplus
PLATFORM -- Linux/x86_64 hfsplus-testing-0001 6.18.0-rc3+ #96 SMP PREEMPT_DYNAMIC Wed Nov 19 12:47:37 PST 2025
MKFS_OPTIONS -- /dev/loop51
MOUNT_OPTIONS -- /dev/loop51 /mnt/scratch
generic/101 32s ... 30s
Ran: generic/101
Passed all 1 tests
sudo fsck.hfsplus -d /dev/loop51
** /dev/loop51
Using cacheBlockSize=32K cacheTotalBlock=1024 cacheSize=32768K.
Executing fsck_hfs (version 540.1-Linux).
** Checking non-journaled HFS Plus Volume.
The volume name is untitled
** Checking extents overflow file.
** Checking catalog file.
** Checking multi-linked files.
** Checking catalog hierarchy.
** Checking extended attributes file.
** Checking volume bitmap.
** Checking volume information.
** The volume untitled appears to be OK.
Signed-off-by: Viacheslav Dubeyko <slava@dubeyko.com>
cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
cc: Yangtao Li <frank.li@vivo.com>
cc: linux-fsdevel@vger.kernel.org
Link: https://lore.kernel.org/r/20251119223219.1824434-1-slava@dubeyko.com
Signed-off-by: Viacheslav Dubeyko <slava@dubeyko.com>
__blkdev_issue_discard() always returns 0, making the error check
in blkdev_issue_discard() dead code.
In function blkdev_issue_discard() initialize ret = 0, remove ret
assignment from __blkdev_issue_discard(), rely on bio == NULL check to
call submit_bio_wait(), preserve submit_bio_wait() error handling, and
preserve -EOPNOTSUPP to 0 mapping.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Chaitanya Kulkarni <ckulkarnilinux@gmail.com>
Reviewed-by: Anuj Gupta <anuj20.g@samsung.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now that all pieces are in place, change the implementations of
sched_mm_cid_fork() and sched_mm_cid_exit() to adhere to the new strict
ownership scheme and switch context_switch() over to use the new
mm_cid_schedin() functionality.
The common case is that there is no mode change required, which makes
fork() and exit() just update the user count and the constraints.
In case that a new user would exceed the CID space limit the fork() context
handles the transition to per CPU mode with mm::mm_cid::mutex held. exit()
handles the transition back to per task mode when the user count drops
below the switch back threshold. fork() might also be forced to handle a
deferred switch back to per task mode, when a affinity change increased the
number of allowed CPUs enough.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172550.280380631@linutronix.de
When affinity changes cause an increase of the number of CPUs allowed for
tasks which are related to a MM, that might results in a situation where
the ownership mode can go back from per CPU mode to per task mode.
As affinity changes happen with runqueue lock held there is no way to do
the actual mode change and required fixup right there.
Add the infrastructure to defer it to a workqueue. The scheduled work can
race with a fork() or exit(). Whatever happens first takes care of it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172550.216484739@linutronix.de
CIDs are either owned by tasks or by CPUs. The ownership mode depends on
the number of tasks related to a MM and the number of CPUs on which these
tasks are theoretically allowed to run on. Theoretically because that
number is the superset of CPU affinities of all tasks which only grows and
never shrinks.
Switching to per CPU mode happens when the user count becomes greater than
the maximum number of CIDs, which is calculated by:
opt_cids = min(mm_cid::nr_cpus_allowed, mm_cid::users);
max_cids = min(1.25 * opt_cids, nr_cpu_ids);
The +25% allowance is useful for tight CPU masks in scenarios where only a
few threads are created and destroyed to avoid frequent mode
switches. Though this allowance shrinks, the closer opt_cids becomes to
nr_cpu_ids, which is the (unfortunate) hard ABI limit.
At the point of switching to per CPU mode the new user is not yet visible
in the system, so the task which initiated the fork() runs the fixup
function: mm_cid_fixup_tasks_to_cpu() walks the thread list and either
transfers each tasks owned CID to the CPU the task runs on or drops it into
the CID pool if a task is not on a CPU at that point in time. Tasks which
schedule in before the task walk reaches them do the handover in
mm_cid_schedin(). When mm_cid_fixup_tasks_to_cpus() completes it's
guaranteed that no task related to that MM owns a CID anymore.
Switching back to task mode happens when the user count goes below the
threshold which was recorded on the per CPU mode switch:
pcpu_thrs = min(opt_cids - (opt_cids / 4), nr_cpu_ids / 2);
This threshold is updated when a affinity change increases the number of
allowed CPUs for the MM, which might cause a switch back to per task mode.
If the switch back was initiated by a exiting task, then that task runs the
fixup function. If it was initiated by a affinity change, then it's run
either in the deferred update function in context of a workqueue or by a
task which forks a new one or by a task which exits. Whatever happens
first. mm_cid_fixup_cpus_to_task() walks through the possible CPUs and
either transfers the CPU owned CIDs to a related task which runs on the CPU
or drops it into the pool. Tasks which schedule in on a CPU which the walk
did not cover yet do the handover themselves.
This transition from CPU to per task ownership happens in two phases:
1) mm:mm_cid.transit contains MM_CID_TRANSIT. This is OR'ed on the task
CID and denotes that the CID is only temporarily owned by the
task. When it schedules out the task drops the CID back into the
pool if this bit is set.
2) The initiating context walks the per CPU space and after completion
clears mm:mm_cid.transit. After that point the CIDs are strictly
task owned again.
This two phase transition is required to prevent CID space exhaustion
during the transition as a direct transfer of ownership would fail if
two tasks are scheduled in on the same CPU before the fixup freed per
CPU CIDs.
When mm_cid_fixup_cpus_to_tasks() completes it's guaranteed that no CID
related to that MM is owned by a CPU anymore.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172550.088189028@linutronix.de
The MM CID management has two fundamental requirements:
1) It has to guarantee that at no given point in time the same CID is
used by concurrent tasks in userspace.
2) The CID space must not exceed the number of possible CPUs in a
system. While most allocators (glibc, tcmalloc, jemalloc) do not
care about that, there seems to be at least some LTTng library
depending on it.
The CID space compaction itself is not a functional correctness
requirement, it is only a useful optimization mechanism to reduce the
memory foot print in unused user space pools.
The optimal CID space is:
min(nr_tasks, nr_cpus_allowed);
Where @nr_tasks is the number of actual user space threads associated to
the mm and @nr_cpus_allowed is the superset of all task affinities. It is
growth only as it would be insane to take a racy snapshot of all task
affinities when the affinity of one task changes just do redo it 2
milliseconds later when the next task changes it's affinity.
That means that as long as the number of tasks is lower or equal than the
number of CPUs allowed, each task owns a CID. If the number of tasks
exceeds the number of CPUs allowed it switches to per CPU mode, where the
CPUs own the CIDs and the tasks borrow them as long as they are scheduled
in.
For transition periods CIDs can go beyond the optimal space as long as they
don't go beyond the number of possible CPUs.
The current upstream implementation adds overhead into task migration to
keep the CID with the task. It also has to do the CID space consolidation
work from a task work in the exit to user space path. As that work is
assigned to a random task related to a MM this can inflict unwanted exit
latencies.
Implement the context switch parts of a strict ownership mechanism to
address this.
This removes most of the work from the task which schedules out. Only
during transitioning from per CPU to per task ownership it is required to
drop the CID when leaving the CPU to prevent CID space exhaustion. Other
than that scheduling out is just a single check and branch.
The task which schedules in has to check whether:
1) The ownership mode changed
2) The CID is within the optimal CID space
In stable situations this results in zero work. The only short disruption
is when ownership mode changes or when the associated CID is not in the
optimal CID space. The latter only happens when tasks exit and therefore
the optimal CID space shrinks.
That mechanism is strictly optimized for the common case where no change
happens. The only case where it actually causes a temporary one time spike
is on mode changes when and only when a lot of tasks related to a MM
schedule exactly at the same time and have eventually to compete on
allocating a CID from the bitmap.
In the sysbench test case which triggered the spinlock contention in the
initial CID code, __schedule() drops significantly in perf top on a 128
Core (256 threads) machine when running sysbench with 255 threads, which
fits into the task mode limit of 256 together with the parent thread:
Upstream rseq/perf branch +CID rework
0.42% 0.37% 0.32% [k] __schedule
Increasing the number of threads to 256, which puts the test process into
per CPU mode looks about the same.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172550.023984859@linutronix.de
The MM CID management has two fundamental requirements:
1) It has to guarantee that at no given point in time the same CID is
used by concurrent tasks in userspace.
2) The CID space must not exceed the number of possible CPUs in a
system. While most allocators (glibc, tcmalloc, jemalloc) do not care
about that, there seems to be at least librseq depending on it.
The CID space compaction itself is not a functional correctness
requirement, it is only a useful optimization mechanism to reduce the
memory foot print in unused user space pools.
The optimal CID space is:
min(nr_tasks, nr_cpus_allowed);
Where @nr_tasks is the number of actual user space threads associated to
the mm and @nr_cpus_allowed is the superset of all task affinities. It is
growth only as it would be insane to take a racy snapshot of all task
affinities when the affinity of one task changes just do redo it 2
milliseconds later when the next task changes its affinity.
That means that as long as the number of tasks is lower or equal than the
number of CPUs allowed, each task owns a CID. If the number of tasks
exceeds the number of CPUs allowed it switches to per CPU mode, where the
CPUs own the CIDs and the tasks borrow them as long as they are scheduled
in.
For transition periods CIDs can go beyond the optimal space as long as they
don't go beyond the number of possible CPUs.
The current upstream implementation adds overhead into task migration to
keep the CID with the task. It also has to do the CID space consolidation
work from a task work in the exit to user space path. As that work is
assigned to a random task related to a MM this can inflict unwanted exit
latencies.
This can be done differently by implementing a strict CID ownership
mechanism. Either the CIDs are owned by the tasks or by the CPUs. The
latter provides less locality when tasks are heavily migrating, but there
is no justification to optimize for overcommit scenarios and thereby
penalizing everyone else.
Provide the basic infrastructure to implement this:
- Change the UNSET marker to BIT(31) from ~0U
- Add the ONCPU marker as BIT(30)
- Add the TRANSIT marker as BIT(29)
That allows to check for ownership trivially and provides a simple check for
UNSET as well. The TRANSIT marker is required to prevent CID space
exhaustion when switching from per CPU to per task mode.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://patch.msgid.link/20251119172549.960252358@linutronix.de
Prepare for the new CID management scheme which puts the CID ownership
transition into the fork() and exit() slow path by serializing
sched_mm_cid_fork()/exit() with it, so task list and cpu mask walks can be
done in interruptible and preemptible code.
The contention on it is not worse than on other concurrency controls in the
fork()/exit() machinery.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172549.895826703@linutronix.de
Reading mm::mm_users and mm:::mm_cid::nr_cpus_allowed every time to compute
the maximal CID value is just wasteful as that value is only changing on
fork(), exit() and eventually when the affinity changes.
So it can be easily precomputed at those points and provided in mm::mm_cid
for consumption in the hot path.
But there is an issue with using mm::mm_users for accounting because that
does not necessarily reflect the number of user space tasks as other kernel
code can take temporary references on the MM which skew the picture.
Solve that by adding a users counter to struct mm_mm_cid, which is modified
by fork() and exit() and used for precomputing under mm_mm_cid::lock.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172549.832764634@linutronix.de
It's getting bigger soon, so just move it out of line to the rest of the
code.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172549.769636491@linutronix.de
There is no need anymore to keep this under sighand lock as the current
code and the upcoming replacement are not depending on the exit state of a
task anymore.
That allows to use a mutex in the exit path.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20251119172549.706439391@linutronix.de
This is truly a bitmap and just conveniently uses a cpumask because the
maximum size of the bitmap is nr_cpu_ids.
But that prevents to do searches for a zero bit in a limited range, which
is helpful to provide an efficient mechanism to consolidate the CID space
when the number of users decreases.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Yury Norov (NVIDIA) <yury.norov@gmail.com>
Link: https://patch.msgid.link/20251119172549.642866767@linutronix.de
