| Age | Commit message (Collapse) | Author |
|---|
|
In order to support preempt_disable()-like interrupt disabling, that is,
using part of preempt_count() to track interrupt disabling nesting
level, change the preempt_count() layout to contain 8-bit
HARDIRQ_DISABLE count.
Signed-off-by: Boqun Feng <boqun@kernel.org>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Boqun Feng <boqun@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260121223933.1568682-2-lyude@redhat.com
Link: https://patch.msgid.link/20260804161447.84806-3-boqun@kernel.org
|
|
Move NMI nesting tracking from the preempt_count bits to a separate
per-CPU counter (nmi_nesting). This is to free up the NMI bits in the
preempt_count, allowing those bits to be repurposed for other uses.
Reduce NMI_BITS from 4 to 1, using it only to detect if we're in an NMI.
The per-CPU counter currently caps nesting at 15.
[boqun: Address Steven Rostedt's comment on the BUG_ON() condition]
[boqun: Use preempt_count_set() in __nmi_exit() to avoid underflow]
Suggested-by: Boqun Feng <boqun@kernel.org>
Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Boqun Feng <boqun@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260121223933.1568682-3-lyude@redhat.com
Link: https://patch.msgid.link/20260804161447.84806-2-boqun@kernel.org
|
|
We currently have two implementations of get_timer offset(), one
in arm_arch_timer.h, and another one in switch.h.
These two only differ by a pair of kern_hyp_va(), which seems a
pretty weak reason to open-code it.
Turn this function into a macro to avoid the include dependency hell
on kern_hyp_va(), and make it work correctly in all contexts.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Mostafa Saleh <smostafa@google.com>
Link: https://patch.msgid.link/20260808085824.732659-2-smostafa@google.com
Signed-off-by: Oliver Upton <oupton@kernel.org>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux into soc/dt
Qualcomm Arm64 DeviceTree updates for v7.3
Introduce DeviceTree support for the Shikra SoC, its GCC and RPM clock
controllers, CQM, CQS, and IQS SoM platforms, and their evaluation boards.
Document the Hawi and Maili SoCs.
Add the Eliza CQS SoM and EVK platforms, the QCS8550 RB5 Gen 2, Vicharak
Axon Mini, HONOR MagicBook Art 14, Microsoft Surface Pro 12, Xiaomi 12 Lite,
Motorola Edge 30, Sony Xperia M2, and Motorola Moto G2 device trees.
Extend Eliza support with USB, SD card, touchscreen, PMIC, interconnect,
thermal, CPU and LLCC bandwidth-monitor, QUPv3, and ADSP GPR descriptions.
Expand Glymur support with GPU, camera and EVA clock controllers, LPASS
audio, CoreSight, PCIe, USB, IMEM and PIL memory regions, power domains,
thermal cooling, and CRD peripherals. Add SoCCP, PMIC regulator, TRNG, and
CPU-capacity descriptions.
Add display, IPA, camera, USB, audio, crypto, TRNG, CoreSight, and thermal
support to the Kaanapali, Milos, Hamoa, and related platforms. Add embedded
controller support for Hamoa and Glymur boards.
Improve networking-platform descriptions with IPQ PCIe port, PHY, clock,
PWM, regulator, and interrupt updates. Update CTCU, display, Iris, audio,
camera, PCIe, USB-C, power-domain, memory-region, and thermal descriptions
across established Qualcomm platforms.
Correct bindings and DeviceTree validation for Qualcomm compatible strings,
legacy fallback compatibles, node naming and formatting, address ranges,
clock specifiers, reserved memory, regulators, GPIOs, and peripheral wiring.
* tag 'qcom-arm64-for-7.3' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux: (244 commits)
arm64: dts: qcom: eliza: Describe the ADSP GPR node
arm64: dts: qcom: eliza-evk: Add support for USB and SD card
dt-bindings: arm: qcom-soc: Allow WSA88xx speaker compatible
dt-bindings: arm: qcom-soc: Validate nodes with fallbacks
dt-bindings: arm: qcom-soc: Document more of existing legacy style compatibles
dt-bindings: arm: qcom-soc: Include Eliza, Kaanapali and others in SoC names
arm64: dts: qcom: eliza: Enable cpufreq cooling devices
arm64: dts: qcom: glymur: add SoCCP DT node
arm64: dts: qcom: glymur: fix SoCCP memory mappings
arm64: dts: qcom: mahua: Add QREF regulator supplies to TCSR
arm64: dts: qcom: glymur: Add QREF regulator supplies to TCSR
arm64: dts: qcom: glymur: Add CX power domain to GCC
arm64: dts: qcom: glymur: Drop fake PCIe phy 3B
arm64: dts: qcom: glymur: add TRNG node
arm64: dts: qcom: glymur: enable ETR and CTCU devices
arm64: dts: qcom: glymur: Add PCIe port compatibles and ports
arm64: dts: qcom: smb2370: Disable SMB2370_2 by default
arm64: dts: qcom: eliza-mtp: Enable touchscreen
arm64: dts: qcom: purwa-iot-som: enable video
arm64: dts: qcom: purwa: Override Iris clocks and operating points
...
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/driver-core/driver-core
Pull driver core fixes from Danilo Krummrich:
- Fix Rust build failure on s390 by gating ioremap() / iounmap()
helpers and the io::mem module on CONFIG_HAS_IOMEM; gate affected
doctests as well.
- Add missing kernel-doc for show_const / store_const union members in
struct device_attribute.
* tag 'driver-core-7.2-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/driver-core/driver-core:
rust: io: gate ioremap doctests on CONFIG_HAS_IOMEM
rust: io: gate ioremap/iounmap on CONFIG_HAS_IOMEM
driver core: add missing kernel-doc for union members
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Pull input updates from Dmitry Torokhov:
- Fixes for information leaks and OOB accesses across several drivers,
including evdev, focaltech, edt-ft5x06, iforce, and cs40l50-vibra
- Improvements to the synaptics-rmi4 driver to properly handle F54
worker errors and prevent buffer overflows
- Input validation fixes in the hynitron_cstxxx touchscreen driver to
prevent issues with invalid finger IDs and touch counts
- Fixes for use-after-free and initialization bugs in the byd mouse and
psxpad-spi drivers
- New quirks for the atkbd driver to make keyboard work on HONOR and
Xiaomi laptops
- Support for the ZENAIM LEVERLESS controller in the xpad driver.
* tag 'input-for-v7.2-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: evdev - sanitize event type index when fetching event masks
Input: synaptics-rmi4 - propagate F54 worker errors to V4L2 queue
Input: synaptics-rmi4 - block s_input when F54 queue is busy
Input: synaptics-rmi4 - bound the F54 report size to the allocated buffer
Input: synaptics-rmi4 - zero report size on F54 work error
Input: synaptics-rmi4 - fix F55 transmitter electrode count typo
Input: hynitron_cstxxx - validate touch count and finger IDs
Input: evdev - fix information leak in evdev_pass_values()
fixp-arith: convert comments to kernel-doc format
Input: focaltech - fix array out-of-bounds in focaltech_process_rel_packet
Input: atkbd - skip deactivate for HONOR ZQC-P
Input: atkbd - skip deactivate for Xiaomi Book Pro 14's internal keyboard
Input: iforce - validate input packet lengths
Input: psxpad-spi - set driver data before use
Input: cs40l50-vibra - validate custom data from user space
Input: xpad - add support for ZENAIM LEVERLESS
Input: edt-ft5x06 - ignore contacts with an out-of-range slot id
Input: byd - synchronize timer deletion before freeing private data
|
|
Implement ->set_speed() so the I3C core can switch open-drain timing for
the first broadcast address per spec: I3C_OPEN_DRAIN_SLOW_SPEED programs
tHIGH_INIT (200 ns) before RSTDAA, and I3C_OPEN_DRAIN_NORMAL_SPEED restores
normal OD timing afterward. Cache the normal OD register value during bus
init and use a separate od_hcnt for the slow path so SDR extended timing
remains derived from the normal PP hcnt.
For AMD_I3C_OD_PP_TIMING, cache AMD_I3C_OD_TIMING as the normal OD
baseline and stop rewriting OD timing in send_ccc_cmd()/runtime resume so
I3C_OPEN_DRAIN_SLOW_SPEED is preserved through RSTDAA.
Use PM_RUNTIME_ACQUIRE_AUTOSUSPEND() in set_speed(). Compute od_hcnt with
DIV_ROUND_UP_ULL() for 32-bit safety and clamp it to U8_MAX to match the
8-bit I3C_OD_HCNT field.
Fixes I2C devices with spike filters not being detected on mixed buses.
Signed-off-by: Tze Yee Ng <tze.yee.ng@altera.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/d789219ca0418898a1ef2bf9295b4f96ca7b4209.1785484707.git.tze.yee.ng@altera.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
Add i3c_master_has_wakeup_enabled_devs(), which iterates over the devices
on an I3C bus and reports whether any of them are enabled for system
wakeup and have IBI enabled.
Controller drivers can use this helper to determine whether wakeup
support must remain available while the system is suspended.
Acked-by : Mukesh Savaliya <mukesh.savaliya@oss.qualcomm.com>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260807145638.168865-11-adrian.hunter@intel.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
An I3C controller acts as a bus controller for one or more I3C devices.
If the controller can wake the system in response to an In-Band
Interrupt (IBI), then any device on that bus that is capable of
generating IBIs can potentially be used as a wakeup source.
Add an ibi_wakeup flag to struct i3c_master_controller so controller
drivers can advertise support for IBI-based wakeup.
If set, mark IBI-capable I3C devices as wakeup capable when they are
registered, allowing wakeup management through the standard device
wakeup framework.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Acked-by: Mukesh Savaliya <mukesh.savaliya@oss.qualcomm.com>
Link: https://patch.msgid.link/20260807145638.168865-9-adrian.hunter@intel.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
i3c_master_register_new_i3c_devs() registers newly discovered devices
while holding i3c_bus_normaluse_lock(), a down_read(). device_register()
can immediately probe the device, and probe callbacks typically invoke
I3C helpers that take i3c_bus_normaluse_lock() again, leading to a
recursive acquisition of the same rwsem. rwsems do not support recursive
read locking and can deadlock when a writer is waiting. See the
"Recursive read locks" section of Documentation/locking/lockdep-design.rst.
For example, with Intel LPSS I3C, LOCKDEP generates a WARNING like:
# echo intel-lpss-i3c.0 > /sys/bus/platform/drivers/mipi-i3c-hci/unbind
# echo intel-lpss-i3c.0 > /sys/bus/platform/drivers/mipi-i3c-hci/bind
WARNING: possible recursive locking detected
kworker/5:1/94 is trying to acquire lock:
ffff88811c810d78 (&i3cbus->lock){++++}-{4:4}, at: i3c_device_match_id+0x45/0x370
but task is already holding lock:
ffff88811c810d78 (&i3cbus->lock){++++}-{4:4}, at: i3c_master_reg_work_fn+0x21/0x5f0
Fix this by separating device creation from device registration.
Populate desc->dev under the maintenance lock, collect the devices that
still need registration into a local list, then release the lock before
calling device_register(). Finally retake the lock and clean up any
devices that failed to register.
Use the maintenance lock rather than the normal-use lock while adding
device objects. A write-side maintenance lock prevents readers from
observing a partially initialized desc->dev during initial device
population, or desc->dev disappearing if registration fails.
The local list requires a list node, so add a list node member to struct
i3c_device.
Fixes: 3a379bbcea0a ("i3c: Add core I3C infrastructure")
Cc: stable@vger.kernel.org
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260807145638.168865-2-adrian.hunter@intel.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
A struct_ops callback cannot receive an arena pointer directly, so
passing one takes two steps. The pointer arrives as a bare u64 that the
callback casts, and because the two sides address the arena through
different bases it also has to be rebased by hand on the way in.
Add the __arena and __arena__nullable stub argument suffixes to make this
convenient. The callback declares the parameter as an arena pointer,
receives it as a PTR_TO_ARENA register, and dereferences it directly,
while the kernel caller just passes the natural kernel arena address
(kaddr). The trampoline converts the value while saving the arguments
into the BPF ctx, ctx[slot] = (u32)(kaddr - kern_vm_start), so the
program never sees a kernel address and nothing rewrites the ctx after
the fact. The converted value keeps the upper 32 bits clear as the JITs
require of arena pointer registers and behaves like any cast_kern'ed
arena pointer, so cast_user recovers the full user-visible address.
__arena converts unconditionally and the kernel caller must not pass
NULL. __arena__nullable preserves NULL, tested on the full 64-bit kernel
pointer, and surfaces to the verifier as PTR_TO_ARENA (but not as a
PTR_TO_ARENA | PTR_MAYBE_NULL). The reason is that PTR_TO_ARENA in the
program's type state already encompasses NULL-ness, so it is not
meaningful to force a NULL check for the program.
The composite suffix intentionally ends in __nullable. Classify
__arena__nullable before the generic suffix so scalar arena pointees do
not take the generic nullable BTF pointer path.
This patch adds the generic side. prepare_arg_info() records arena and
nullable argument flags in the struct_ops function model, and
bpf_tramp_arena_base() returns the arena base for a single-program
struct_ops indirect trampoline. Only that trampoline converts: its
program's arena is fixed at generation time. Generic trampolines can mix
programs with different arenas and reject arena context arguments
defensively, which is unreachable today as only struct_ops programs
carry them. Architectures that do not implement the conversion are
gated out at verification time with bpf_jit_supports_arena_args().
Co-developed-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://patch.msgid.link/20260808003938.3486067-6-memxor@gmail.com
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
|
|
Passing an arena pointer to a kfunc takes two steps today. There is no
arena pointer argument type, so the pointer crosses the boundary as a
bare scalar, and the kfunc then offsets it by the arena base and casts
it before it can touch the memory. Every such kfunc open-codes the same
translation.
Add the __arena and __arena__nullable argument suffixes to make this more
convenient. The kfunc declares the parameter by its real pointer type
and dereferences it directly, with the JIT rebasing the value at the
call site, rN = kern_vm_start + (u32)rN. No bounds check is needed: the
u32 offset stays within the guard-padded arena kernel mapping, and a
fault on an unpopulated page recovers through the per-arena scratch
page. A suffixed argument accepts a PTR_TO_ARENA or scalar register,
matching global subprog arena arguments.
__arena rebases unconditionally, so the kfunc never sees NULL and a
value with zero in the low 32 bits arrives as the arena base.
__arena__nullable preserves NULL for optional arguments by skipping the
rebase when the truncated value, arena offset 0, is zero. Keeping the
plain form NULL-free saves the NULL test on every call.
The double separator makes the annotations composable:
__arena__nullable also ends in __nullable and naturally follows the
common nullable argument path. Plain __arena follows that path too for
verifier type checking because both forms accept a constant zero; the
function-model flag still determines whether the JIT preserves NULL or
rebases it to the arena base.
This patch adds the verifier side: the suffixes are recognized in
check_kfunc_args() and distilled into argument flags in the function
model stored in the kfunc descriptor. JITs retrieve the model while
emitting the call, avoiding per-call state in insn_aux_data.
JITs declare support with bpf_jit_supports_arena_args() and verification
fails with -ENOTSUPP elsewhere.
Co-developed-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://patch.msgid.link/20260808003938.3486067-5-memxor@gmail.com
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
|
|
BTF processing is split around subprogram discovery. The first phase gets
program BTF and imports func_info because a BTF-tagged exception callback
may not be referenced by any instruction. Subprogram discovery needs this
metadata to find it.
The later phase validates func_info and line_info against the complete
subprogram table and applies CO-RE relocations. This split breaks a real
dependency cycle rather than merely running the same checks early.
Rename bpf_check_btf_info_early() and check_btf_func_early() to preparation
names that reflect this role. Add short call-site comments to make the two
phases and their responsibilities clear.
No functional change is intended.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Reviewed-by: Amery Hung <ameryhung@gmail.com>
Link: https://patch.msgid.link/20260808003938.3486067-2-memxor@gmail.com
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
|
|
As reported in the thread [1], the verifier's 32-bit operations zero
extension logic is broken. This logic is responsible for correct
semantics of 32-bit operations on s390 architecture.
According to BPF semantics, operation `w1 += 1` is supposed to zero
extend the upper half of the register `r1`. On s390 the JIT relies on
the verifier emitting explicit zero extension before such operations.
The verifier attempts to minimize the amount of zero extensions
inserted by tracking whether upper halves of the 64-bit registers are
ever used. Previously such tracking worked as follows:
- bpf_reg_state->subreg_def field was set by do_check_insn()
for each operation defining lower but not the upper halves
of the register.
- Whenever an operation reading the whole register was verified,
the verifier checked register's subreg_def and set
bpf_insn_aux_data->zext_dst flag as true via a call to
mark_insn_zext() function.
- After the verification was complete, a special pass
bpf_opt_subreg_zext_lo32_rnd_hi32() extended 32-bit operations
with bpf_insn_aux_data->zext_dst set as true by adding
explicit zero extension.
Note that the logic above relies on bpf_reg_state->subreg_def,
which is a property of a current verifier state.
Before the commit [2] two additional steps happened:
- The verifier tracked upper and lower register halves' liveness as
flags REG_LIVE_READ{32,64} in bpf_reg_state->live.
- The function propagate_liveness() called mark_insn_zext()
in order to transfer the knowledge about which registers have
their upper halves alive (and thus might require zero extension).
The commit [2] removed the two steps described above,
hence making possible a situation like below:
- The register's upper half is set and is used on some verification
path P1 and the register happens not to be marked as precise.
- The checkpoint C is created while processing some instruction
between register initialization and usage.
- On some other verification path P2 the register's upper half is not
initialized and that path ends hitting the checkpoint C.
- In such a case the register's initialization on path P2 would lack
zext_dst mark, making it possible for the program to inject
an arbitrary value in the register's upper half.
This commit replaces subreg_def based logic with computing zext_dst
statically, as a part of the bpf_compute_live_registers() analysis:
- The analysis now tracks usage of upper and lower halves of the
registers separately.
- If some instruction defines a 32-bit subregister, but not the whole
register, *and* the upper half of the register is alive after that
instruction, the instruction is marked as zext_dst.
There is one notable drop in precision: whenever a BPF subprogram is
called, all 64 bits of parameter registers are presumed to be used.
The assumption is that such a drop in precision would not inflict
a noticeable performance penalty.
[1] https://lore.kernel.org/bpf/CAGKGUv=sOuqQtA1Ub-5JXfA4FPosJFYKAQE4B79cK+P1erxqtg@mail.gmail.com/
[2] commit 107e16979905 ("bpf: disable and remove registers chain based liveness")
Fixes: 107e16979905 ("bpf: disable and remove registers chain based liveness")
Reported-by: Min-gyu Kim <gimm78064@gmail.com>
Reported-by: STAR Labs SG <info@starlabs.sg>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/CAGKGUv=sOuqQtA1Ub-5JXfA4FPosJFYKAQE4B79cK+P1erxqtg@mail.gmail.com/
Link: https://lore.kernel.org/bpf/20260807-static-zext-v4-5-b6c270013c77@gmail.com
|
|
https://gitlab.freedesktop.org/drm/misc/kernel into drm-next
drm-misc-next for v7.3:
UAPI Changes:
- Remove the default udmabuf size limit of 64MB.
Cross-subsystem Changes:
- Add dmemcg support for eviction, and hook it up for amdgpu and xe.
Core Changes:
- Changes to TTM to be more aggressive when allocating below protection limit!
- Improve dt binding documentation for renesas.
- Add helper to convert physical address back to buddy block,
add that to and improve its kunit test.
Driver Changes:
- Assorted small fixes to ti-sn65dsi86, panthor, imagination, omapdrm,
bridge/synopsys, panel-edp, ssd130x, panel/tdo-tl070wsh30.
- Add Sharp LQ120P1JX51 panel.
- Add dmemcg support to nouveau.
- Various updates and improvements to sun4i, among which YUV and 4k support.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patch.msgid.link/917d462a-8976-4a15-bec4-4513ec51c5c0@linux.intel.com
|
|
Every devmem dmabuf binding today hands the page_pool PAGE_SIZE niovs.
This caps a single RX descriptor at PAGE_SIZE, burning CPU on buffer
churn for large flows.
Add a bind-time netlink attribute, NETDEV_A_DMABUF_RX_PAGE_SIZE, that
lets userspace request a larger niov size. The value must be a power of
two >= PAGE_SIZE.
The TX path is changed to always pass PAGE_SIZE.
Measurements:
Setup: kperf in devmem RX/TX cuda mode, 4 flows, 64 MB messages, 60s,
dctcp, num-rx-queues=4, dmabuf-rx/tx-size-mb=2048, 10 runs per niov
size, mlx5.
CPU Util:
niov net sirq % net idle % app sys % app idle %
----- ---------------- ---------------- ---------------- ----------------
4K 62.38 +/- 8.27 33.40 +/- 7.51 54.15 +/- 10.23 43.67 +/- 10.53
16K 58.91 +/- 5.35 35.23 +/- 5.88 41.05 +/- 8.87 56.42 +/- 9.24
32K 64.12 +/- 0.68 31.09 +/- 1.48 44.54 +/- 3.51 52.63 +/- 3.65
64K 54.69 +/- 5.54 39.67 +/- 5.81 35.47 +/- 3.11 61.97 +/- 3.27
RX app sys % drops ~19% from 4K to 64K.
Throughput:
niov RX dev Gbps RX flow avg Gbps
----- ---------------- -----------------
4K 300.63 +/- 53.21 75.16 +/- 13.30
16K 321.35 +/- 28.20 80.34 +/- 7.05
32K 347.63 +/- 2.20 86.91 +/- 0.55
64K 332.11 +/- 14.26 83.03 +/- 3.56
Throughput seems to increase, but the stdev is pretty wide so could just
be noise.
kperf support (not yet merged):
https://github.com/facebookexperimental/kperf/commit/8837577f920876bce6986ec18869ac04439ebcd2
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
Link: https://patch.msgid.link/20260805-tcpdm-large-niovs-v8-1-3e0225e2808c@meta.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
To support up to 8 packets per CQE, update related CQE processing
code and structures.
Update ethtool handlers to set this feature.
Update per queue stat to show the coalesced CQE counters.
This feature is supported on NIC hardware showing the relevant
PF flag.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20260805185404.1052177-1-haiyangz@linux.microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This patch moves the check for available free space for a new entry into
a separate function. Existing callers that only check for a non-zero
return value are unaffected. __ptr_ring_produce() now returns -EINVAL
for a zero-size ring and -ENOSPC when full, whereas before both cases
returned -ENOSPC. The new helper allows callers to determine in advance
whether a single subsequent __ptr_ring_produce() call will succeed. This
information can, for example, be used to temporarily stop producing until
__ptr_ring_check_produce() indicates that space is available again.
The return values are documented above the helper, as a caller that waits
for space must distinguish the transient -ENOSPC from the permanent
-EINVAL.
Co-developed-by: Tim Gebauer <tim.gebauer@tu-dortmund.de>
Signed-off-by: Tim Gebauer <tim.gebauer@tu-dortmund.de>
Signed-off-by: Simon Schippers <simon.schippers@tu-dortmund.de>
Link: https://patch.msgid.link/20260803183641.96882-5-simon.schippers@tu-dortmund.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add tun_wake_queue() to tun.c and export it for use by vhost-net. The
function validates that the file belongs to a device implemented by
drivers/net/tun.c, in IFF_TUN as well as in IFF_TAP mode, and that the
tfile exists, dereferences the tun_struct under RCU, and delegates to
__tun_wake_queue().
vhost_net_buf_produce() now calls tun_wake_queue() after a successful
batched consume of the ring to allow the netdev subqueue to be woken up.
The point is to allow the queue to be stopped when it gets full, which is
required for traffic shaping, implemented by the following
"stop tail-drop when IFF_BACKPRESSURE is set".
As __tun_wake_queue() returns early unless IFF_BACKPRESSURE is set, a
tun/tap device that does not opt in only pays for the added check.
macvtap and ipvtap rings, which get_tap_ptr_ring() accepts too, are
unaffected: their producer is the tap_handle_frame() rx_handler and not
ndo_start_xmit, so stopping a netdev TX queue would not hold it back.
drivers/net/tap.c has no netdev_ops of its own either. No
tap_wake_queue() is needed.
cons_cnt and the wake decision are best-effort and are not reverted by
ptr_ring_unconsume(), so vhost_net_buf_unproduce() can leave the subqueue
woken over a full ring. The producer re-stops it on the next packet, and
that path only runs from vhost_net_stop_vq() and vhost_net_set_backend(),
when the consumer is going away, so a stopped queue is the correct end
state rather than a stall.
Co-developed-by: Tim Gebauer <tim.gebauer@tu-dortmund.de>
Signed-off-by: Tim Gebauer <tim.gebauer@tu-dortmund.de>
Signed-off-by: Simon Schippers <simon.schippers@tu-dortmund.de>
Link: https://patch.msgid.link/20260803183641.96882-4-simon.schippers@tu-dortmund.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add the IFF_BACKPRESSURE flag to the UAPI header and to its tools/ copy.
The flag has no effect yet, it is the opt-in switch for the qdisc
backpressure logic added by the following patches.
It is added to TUN_FEATURES only in the last patch of the series, once the
implementation is complete. Until then TUNSETIFF silently masks it off, as
it does for any flag outside TUN_FEATURES.
Keeping the flag and its users in separate patches would either leave a
window where backpressure is unconditional, or make the opt-in a later
add-on. Adding the flag first lets every following patch be a no-op
unless it is set.
Signed-off-by: Simon Schippers <simon.schippers@tu-dortmund.de>
Link: https://patch.msgid.link/20260803183641.96882-2-simon.schippers@tu-dortmund.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add ndisc_check_ns_na(), a standalone NS/NA packet validator modeled
after ipv6_mc_check_mld(). It performs the RFC 4861 section 7.1.1
(Neighbor Solicitation) and 7.1.2 (Neighbor Advertisement) mandatory
checks that are relevant for software operating at the bridge level,
where packets bypass the normal IPv6 stack path:
- Hop Limit must be 255 (packet was not forwarded by a router)
- ICMPv6 checksum is valid
- ICMP Code is 0
- ICMP length is at least 24 octets (sizeof(struct nd_msg))
- Target Address must not be a multicast address
- All included options have a length that is greater than zero
- NS/DAD: destination must be a solicited-node multicast address
- NS/DAD: no Source Link-Layer Address option when source is unspecified
- NA: Solicited flag must be 0 when IP Destination is multicast
On success the function sets the skb transport header and returns 0,
matching the convention of ipv6_mc_check_mld().
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Link: https://patch.msgid.link/20260803112505.613873-3-danieller@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When cookie authentication is disabled, COOKIE_ECHO restores fixed-size
AUTH fields directly from peer-controlled cookie bytes. A forged RANDOM
length, HMAC list, or CHUNKS list can then reach association consumers
with lengths or identifiers that were never validated against the local
backing arrays.
A forged RANDOM length can cause out-of-bounds reads during key-vector
construction. A forged HMAC identifier also caused a 32-byte write past
a zero-length AUTH chunk, providing a primitive for a local privilege
escalation chain.
Validate the cookie's RANDOM, HMACS, and CHUNKS parameters at the cookie
trust boundary before copying them into the association. Reject invalid
types, malformed lengths, unsupported HMAC identifiers, HMAC lists
without SHA1, and forbidden chunk ids.
Fixes: bbd0d59809f9 ("[SCTP]: Implement the receive and verification of AUTH chunk")
Fixes: 1f485649f529 ("[SCTP]: Implement SCTP-AUTH internals")
Signed-off-by: Jérémy Jean <Jeremy.Jean@oss.cyber.gouv.fr>
Acked-by: Xin Long <lucien.xin@gmail.com>
Link: https://patch.msgid.link/20260804200042.2412009-1-Jeremy.Jean@oss.cyber.gouv.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Cross-merge BPF and other fixes after downstream PR.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
|
|
hci_event_func() validates skb->len against ev->max_len from the
entry in hci_ev_table[]. By then, the header has already been
stripped by skb_pull(). So the max event payload is 255, but
hci_ev_table[] still uses HCI_MAX_EVENT_SIZE (260) for it, which is
imprecise.
Fix by introducing HCI_MAX_EVENT_PLEN (255) and using it instead.
Signed-off-by: Zijun Hu <zijun.hu@oss.qualcomm.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
Introduce the hook to solve issues below:
msft_vendor_evt(), the current handler for all VSEs, is unsuitable
since:
- many VSEs are not MSFT ones;
- it always corrupts the non-MSFT VSEs by calling skb_pull_data()
once the MSFT extension is enabled.
Several issues are caused by many transport drivers pre-processing
VSEs in their RX path, often an IRQ-disabled atomic context. Take
the two typical cases below as examples:
Case 1:
// no btmon log, no way to reach userspace
Step 1: handle and free @original_skb directly
Case 2:
// hurts performance and consumes GFP_ATOMIC memory
Step 1: cloned_skb = skb_clone(original_skb, GFP_ATOMIC);
// the VSE is handled here
Step 2: handle and free @cloned_skb
Step 3: hci_recv_frame(hdev, original_skb);
// already handled, but re-enters the stack's event-handling path
Step 4: hci_event_packet(hdev, original_skb);
Fix by introducing the hook with usage:
1) the transport driver registers the hook for VSEs of interest;
2) the stack calls it in process context, handling the VSE like any
other event:
- if interested, handle the VSE - no need to free it - and
return true;
- otherwise return false.
Signed-off-by: Zijun Hu <zijun.hu@oss.qualcomm.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
hci_reset_dev() injects a constant hardware error code 0x00 to restart
the device. But a transport driver may need a different error code.
Fix by introducing __hci_reset_dev(hdev, hw_err_code), which will be
used by a follow-up patch.
Signed-off-by: Zijun Hu <zijun.hu@oss.qualcomm.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
To separate the coredump header and data far more easily, give a
vendor driver the option to pad its header to a fixed size, by
moving the header size limit and ending marker to coredump.h:
- HCI_DEVCD_HDR_SIZE_MAX: the max header size
- HCI_DEVCD_HDR_END_MARKER: the header-ending marker
Signed-off-by: Zijun Hu <zijun.hu@oss.qualcomm.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
Add context analysis annotations for hci_conn::l2cap_data locking.
Also add necessary lockdep_assert_held() and __must_hold annotations
to prove the access is safe.
The access in smp_conn_security() is supposed to be guarded by the
caller holding lock that blocks concurrent l2cap_conn_del() eg.
hdev->lock, conn->lock or chan->lock. Mark unsafe as can't be
automatically checked now.
Signed-off-by: Pauli Virtanen <pav@iki.fi>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
Add MGMT_OP_LOAD_CONN_SUBRATE (0x005C) command to load per-device
connection subrate parameters when the SCI feature is supported.
Add MGMT_EV_CONN_SUBRATE (0x0033) event to notify userspace when
connection rate changes occur via the LE Connection Rate Change HCI
event.
Add subrate fields (subrate_min, subrate_max, max_latency, cont_num)
to struct hci_conn_params to store the loaded subrate parameters, and
the corresponding le_rate_* fields to struct hci_conn to track the
parameters currently in use.
When a single entry is loaded for an already-connected central, or on
connection completion, the LE Connection Rate Request procedure is
initiated to apply the parameters.
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
Add MGMT_SETTING_SCI (bit 25) to advertise support for the Shorter
Connection Interval (SCI) feature. It is reported in the supported
settings whenever the controller is SCI capable, and in the current
settings whenever LE is enabled and the controller is SCI capable
(SCI has no separate enable command, so it is a passive capability).
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
Add HCI command, event and feature bit definitions for the Bluetooth
6.2 Shorter Connection Interval feature:
Commands:
- HCI_OP_LE_CONN_RATE (0x20a1) - Connection Rate Request
- HCI_OP_LE_SET_DEF_RATE (0x20a2) - Set Default Rate Parameters
- HCI_OP_LE_READ_CONN_INTERVAL (0x20a3) - Read Min Supported
Connection Interval
Events:
- HCI_EVT_LE_CONN_RATE_CHANGE (0x37) - Connection Rate Change
Feature bits:
- HCI_LE_SCI - Shorter Connection Intervals
- HCI_LE_SCI_HOST - Shorter Connection Intervals (Host Support)
During controller init, when SCI is supported:
- Set Shorter Connection Intervals (Host Support) feature via
LE Set Host Feature
- Read Minimum Supported Connection Interval
- Set Default Rate Parameters
The Connection Rate Change event handler updates the connection
interval, latency and supervision timeout on the hci_conn.
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
Introduce both helpers for ACL packet since:
both core and transport drivers extract the handle and data length
from its header in several places.
Both will be used later.
Signed-off-by: Zijun Hu <zijun.hu@oss.qualcomm.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
Introduce hci_devcd_state_name() to describe the devcoredump state by a
string name instead of a plain number, for several reasons:
1) Applying it in coredump.c makes the devcoredump state in log messages
more readable than a plain number.
2) Transport drivers may need to show the devcoredump state name too.
3) In future, the universal state name could be notified to userspace
via uevent, allowing a universal application (e.g. a daemon) to be
developed to save the coredump, which is otherwise discarded by the
device coredump core after 5 minutes (DEVCD_TIMEOUT); see
nxp_coredump_notify().
Also drop a trailing space from two bt_dev_dbg() format strings while
applying it in coredump.c.
Signed-off-by: Zijun Hu <zijun.hu@oss.qualcomm.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
- Avoid using deep idle states during initialization in the intel_idle
driver to work around device handling issues (Rafael Wysocki)
- Fix and refactor the ACPI processor driver code related to ACPI _LPI
support and add ACPI _LPI support to intel_idle based on that ACPI
processor driver update (Rafael Wysocki)
* pm-cpuidle:
intel_idle: Avoid using deep idle states during initialization
intel_idle: Update documentation after adding ACPI _LPI support
intel_idle: Add ACPI _LPI support
intel_idle: Prepare for adding ACPI _LPI support
ACPI: processor: idle: Add switch for strict _LPI processing
ACPI: processor: idle: Relocate acpi_processor_extract_lpi_info()
ACPI: processor: idle: Introduce acpi_processor_extract_lpi_info()
ACPI: processor: idle: Introduce too_many_states() for _LPI
ACPI: processor: idle: Rework flatten_lpi_states()
ACPI: processor: idle: Rearrange loop in acpi_processor_get_lpi_info()
ACPI: processor: idle: Drop redundant _LPI presence checks
ACPI: processor: idle: Rework first-level _LPI states processing
ACPI: processor: idle: Rearrange acpi_processor_get_lpi_info()
ACPI: processor: idle: Introduce lpi_state_debug()
ACPI: processor: idle: Split acpi_processor_evaluate_lpi()
ACPI: processor: idle: Rearrange acpi_processor_evaluate_lpi()
ACPI: processor: idle: Unify debug in acpi_processor_evaluate_lpi()
ACPI: processor: idle: Ignore _LPI states with SYSTEMIO entry method
ACPI: processor: idle: Expand _LPI package sanity checks
|
|
Merge an ACPI processor driver update related to ACPI _LPI support and
the introduction of ACPI _LPI suppor to intel_idle based on that update
for 7.3-rc1.
* intel-idle-lpi:
intel_idle: Update documentation after adding ACPI _LPI support
intel_idle: Add ACPI _LPI support
intel_idle: Prepare for adding ACPI _LPI support
ACPI: processor: idle: Add switch for strict _LPI processing
ACPI: processor: idle: Relocate acpi_processor_extract_lpi_info()
ACPI: processor: idle: Introduce acpi_processor_extract_lpi_info()
ACPI: processor: idle: Introduce too_many_states() for _LPI
ACPI: processor: idle: Rework flatten_lpi_states()
ACPI: processor: idle: Rearrange loop in acpi_processor_get_lpi_info()
ACPI: processor: idle: Drop redundant _LPI presence checks
ACPI: processor: idle: Rework first-level _LPI states processing
ACPI: processor: idle: Rearrange acpi_processor_get_lpi_info()
ACPI: processor: idle: Introduce lpi_state_debug()
ACPI: processor: idle: Split acpi_processor_evaluate_lpi()
ACPI: processor: idle: Rearrange acpi_processor_evaluate_lpi()
ACPI: processor: idle: Unify debug in acpi_processor_evaluate_lpi()
ACPI: processor: idle: Ignore _LPI states with SYSTEMIO entry method
ACPI: processor: idle: Expand _LPI package sanity checks
|
|
Merge cpufreq updates for 7.3-rc1:
- Minor fixes and cleanups in assorted cpufreq drivers (Dan Carpenter,
Guru Das Srinagesh, Haoxiang Li, Karl Mehltretter, Sasha Finkelstein,
and Pan Chuang)
- Fix cpufreq table creation and bios_limits() callback in the Rust
bindings (Priya Bala Govindasamy)
- Add IPQ5210 support to qcom-nvmem driver (Varadarajan Narayanan)
- Adjust the .adjust_perf() cpufreq driver callback to allow the
maximum performance value to be passed to drivers and update the
intel_pstate driver to use it (Rafael Wysocki)
- Set policy->cur to the actual requested frequency in the intel_pstate
driver when the performance policy is used (Rafael Wysocki)
- Simplify HWP handling on Broadwell processors in intel_pstate (Rafael
Wysocki)
- Fix setting minimum P-state at init time in intel_pstate (Rafael
Wysocki)
- Consolidate frequency values computation in intel_pstate and clean up
code in that driver (Rafael Wysocki)
- Add missing kernel-doc desciptions for structure and union members in
the amd-pstate driver (David Vernet)
- Handle missing policy in dynamic EPP callbacks in the amd-pstate
driver (EDAMAMEX)
- Introduce EXPORT_SYMBOL_FOR_PSTATE_UT() to export amd-pstate driver
symbols to the amd-pstate-ut subdriver (K Prateek Nayak)
- Add dynamic EPP as an "energy_performance_preference" mode in
amd-pstate, remove the "amd_dynamic_epp" kernel command line option
and the "dynamic_epp" sysfs attribute, and update the dynamic_epp
documentation accordingly (K Prateek Nayak)
- Add unit tests for CPPC Performance Priority and the "dynamic" EPP
mode in the amd-pstate driver (K Prateek Nayak)
- Set min_limit_freq based on bios_min_perf in amd-pstate and remove
the defensive check for bios_min_perf from it (K Prateek Nayak)
- Fix EPP return type and handle errors in amd-pstate during
initialization, toggle auto_sel in active mode on shared memory
systems, and cache the firmware programmed EPP value (Marco Scardovi)
- Skip tests in amd-pstate-ut if the amd-pstate driver is not in active
use (Qianheng Peng)
- Replace sprintf() with sysfs_emit() in sysfs show in the cpufreq
schedutil governor and fix a self-contradictory comment in
sugov_iowait_apply() (Zhongqiu Han)
- Fix the usage example for the sampling_rate tunable of the ondemand
cpufreq governor in admin-guide (wangxiaodong)
* pm-cpufreq: (40 commits)
cpufreq: imx6q: fix out-of-bounds write when probed more than once
cpufreq: imx6q: fix devres accumulation across driver rebind
rust: cpufreq: Fix temporary write in Registration::bios_limit_callback
rust: cpufreq: Add CPUFREQ_TABLE_END as last table entry in TableBuilder::to_table
cpufreq: intel_pstate: Adjust policy->cur in active mode to policy
cpufreq/amd-pstate: Document missing kernel-doc members
cpufreq/amd-pstate-ut: Add unit test for CPPC Performance Priority
cpufreq/amd-pstate-ut: Add unit test for "dynamic" EPP mode
cpufreq/amd-pstate: Reduce the scope of exported symbols
Documentation/amd-pstate: Update dynamic_epp documentation with new behavior
cpufreq/amd-pstate: Remove "amd_dynamic_epp" cmdline and "dynamic_epp" sysfs
cpufreq/amd-pstate: Add dynamic EPP as an "energy_performance_preference" mode
cpufreq/amd-pstate: Extract platform profile to EPP conversion into a helper
cpufreq/amd-pstate: Remove the defensive check for bios_min_perf
cpufreq/amd-pstate: Set min_limit_freq based on bios_min_perf
cpufreq: apple-soc: Calculate frequency as a 64-bit value
kselftest: cpufreq: Backup and restore governor for sptests
selftests/cpufreq: Remove unnecessary sudo from quick_shuffle()
selftests/cpufreq: Remove unused local variables from switch_show_governor()
cpufreq/amd-pstate: handle missing policy in dynamic EPP callbacks
...
|
|
Merge irqchip changes related to ACPI support that are needed to enable
GICv5 IWB ACPI probe ordering detection on ARM, including the addition
of acpi_device_clear_deps() and RISC-V interrupt controller management
code refactoring to extract generic code from it into the common ACPI
IRQ code (Lorenzo Pieralisi)
* acpi-irqchip:
irqchip/gic-v5: Enable GICv5 IWB ACPI probe ordering detection
ACPI/IORT: Implement ACPI infrastructure to enable GICv5 IWB probe deferral
ACPI: irq: Move RISC-V interrupt controllers autodep to ACPI IRQ code
ACPI: RISC-V: Fix riscv_acpi_add_prt_dep() loop handling
ACPI: RISC-V: Check acpi_get_handle() status in riscv_acpi_add_prt_dep()
ACPI: RISC-V: Fix riscv_acpi_irq_get_dep() loop termination
ACPI: Add acpi_device_clear_deps() helper function
|
|
Merge and ACPI support update related to NUMA, miscellaneous updates
related to ACPI support, and PNP updates for 7.3-rc1:
- Remove a node_set() call that is redundant from
acpi_parse_memory_affinity() (Sang-Heon Jeon)
- Prevent kernel-doc warnings by converting 2 function description
comments to kernel-doc format (Randy Dunlap)
- Fix docs build error in the ACPI admin-guide documentation (Randy
Dunlap)
- Replace __get_free_page() with kmalloc() in the code handling ACPI
NVS memory during system suspend/resume (Mike Rapoport)
- Fix card device cleanup on registration failure in the core PNP
code (Yuho Choi)
- Dropi an unused assignment of pnp_device_id driver data (Uwe
Kleine-König)
* acpi-numa:
ACPI: NUMA: remove redundant node_set() call
* acpi-misc:
ACPI: pmtmr: Convert to kernel-doc format
docs: ACPI: DSD: motorcomm: fix docs build error
ACPI: NVS: replace __get_free_page() with kmalloc()
* pnp:
PNP: Fix card device cleanup on registration failure
PNP: Drop unused assignment of pnp_device_id driver data
|
|
Several HWRM commands carry __le64 DMA address fields in their input
structures; firmware reads from or writes to the memory those addresses
point to.
Have a static per-command descriptor table in the driver that records
the details of the DMA fields in each supported HWRM input struct. When
a DMA-bearing HWRM command arrives, the driver reads the userspace
pointer out of each declared address field and clears the field, allocates
a DMA-coherent kernel buffer sized from the command's own length
information, copies data to/from the userspace pointer, and patches the
field with the real DMA bus address before the command is sent to firmware.
Responses are copied back to the original userspace pointer afterward.
Scope-gated allow-list and timeout value list are updated with the new
the commands.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20260807125846.45570-3-pavan.chebbi@broadcom.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Since bnxt_fwctl is going to support additional commands in the next
patch, add their missing definitions from the firmware spec.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20260807125846.45570-2-pavan.chebbi@broadcom.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
We are using atomic types for the log_commit array of struct btrfs_root
but all we need is simple booleans. The log_commit array elements are
always protected by the root's log_mutex, both for writes and reads, so
we can use a simple boolean. The use of atomics if from the very early
days of the log tree code where the access to the fields was not protected
by any lock.
So switch to simple booleans, which results in cheaper code and slightly
reduces the object size too.
Reviewed-by: Boris Burkov <boris@bur.io>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Several vendor drivers have a requirement to send a synchronous raw HCI
reset with HCI_INIT_TIMEOUT.
Add a dedicated __hci_reset_sync() for them to use.
Signed-off-by: Zijun Hu <zijun.hu@oss.qualcomm.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
hci_cmd_sync_dequeue_once() had a single in-tree caller,
hci_cancel_connect_sync(), which now holds cmd_sync_work_lock across the
in-flight create flag test and the dequeue and so open-codes the lookup
and cancel under that lock. That leaves the exported
hci_cmd_sync_dequeue_once() with no in-tree user, so remove it along with
its declaration.
Signed-off-by: Siwei Zhang <oss@fourdim.xyz>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
|
|
While the GCC and Clang compilers already define __ASSEMBLER__
automatically when compiling assembly code, __ASSEMBLY__ is a
macro that only gets defined by the Makefiles in the kernel.
This can be very confusing when switching between userspace
and kernelspace coding, or when dealing with uapi headers that
rather should use __ASSEMBLER__ instead. So let's standardize now
on the __ASSEMBLER__ macro that is provided by the compilers.
This is a completely mechanical patch (done with a simple "sed -i"
statement).
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260619155733.58852-1-thuth@redhat.com
|
|
Commit 9c63e84db29b ("sched/core: Disable SD_PREFER_SIBLING on asymmetric
CPU capacity domains") removed the SD_PREFER_SIBLING from the domains with
asymmetric capacity. This was done to avoid spreading tasks to sibling
scheduling groups with less capacity, but this does not happen: checks for
capacity in update_sd_pick_busiest(), sched_balance_find_src_group(), and
sched_balance_find_src_rq() prevent migrations from high- to low-capacity
CPUs if the busiest group is not overloaded.
The cluster topology is a notable example: some systems have scheduling
domains spanning CPUs of asymmetric capacity, grouped into two or more
equal-capacity clusters sharing an L2 cache. When CONFIG_SCHED_CLUSTER is
enabled, SD_PREFER_SIBLING is needed in the domain to spread load across
these clusters.
CPUs with spare capacity, big or small, have always helped overloaded
groups. Once the overloading condition disappears, misfit load will still
be used to move high-utilization tasks to bigger CPUs if they have spare
capacity.
Adding the SD_PREFER_SIBLING flag shifts load balancing in shared-LLC
domains from equalizing the number of idle CPUs to equalizing the number
of running tasks. This enables migrations among clusters from newly-idle
load balance, where the outgoing task is already dequeued but the CPU
has not yet transitioned to idle.
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Andrea Righi <arighi@nvidia.com>
Link: https://patch.msgid.link/20260720-rneri-fix-cas-clusters-v6-6-bb500bf4afd4@linux.intel.com
|
|
queued_spin_unlock() gates its contended_release trace call behind a
static branch, so a NOP sits on the unlock path even while the
tracepoint is disabled. Removing that requires replacing the unlock
implementation only while contended_release is enabled, which needs a
callback when the tracepoint is toggled.
Convert contended_release to TRACE_EVENT_FN() and add weak no-op
arch_contended_release_trace_reg()/arch_contended_release_trace_unreg()
hooks.
The default hooks are empty, so this is a no-op until an architecture
overrides them.
No functional change intended.
Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juergen Gross <jgross@suse.com>
Link: https://patch.msgid.link/1c2fcccfb584c075c02890c484f22c76a1948bf1.1785778551.git.d@ilvokhin.com
|
|
Unlike mutex and rw_semaphore, qspinlock has no owner field, so "perf
lock contention --lock-owner" cannot attribute a contended spinlock to
its holder. The waiter-side contention_begin event records that a
spinlock is contended, but not by whom. Firing contended_release in the
holder's context at unlock is the only way to capture the holder of a
contended spinlock.
Combine the contention check, trace call and release in an out-of-line
queued_spin_release_traced() so the compiler need not preserve the lock
pointer in a callee-saved register across the call.
The check in queued_spin_unlock() is paid on every unlock, even while
the tracepoint is disabled: a static-branch NOP on x86_64, and a few
more instructions to manage a stack frame elsewhere. Gate it behind
CONFIG_QUEUED_SPINLOCKS_TRACE_CONTENDED_RELEASE (default n) so nobody
pays for a tracepoint they do not use. Sleeping locks fire
contended_release regardless.
On x86 this generic path is used only with PARAVIRT_SPINLOCKS=n (e.g.
defconfig). PARAVIRT_SPINLOCKS=y kernels keep the paravirt static_call
unlock and are wired up separately.
All below are with the QUEUED_SPINLOCKS_TRACE_CONTENDED_RELEASE option
enabled.
_raw_spin_unlock(), x86_64 defconfig, GCC 11, tracepoint compiled in but
disabled. The unlock is the single 'movb'. The only instruction added to
the executed path is the 2-byte static-branch NOP. The CALL to the
traced helper and the JMP back are emitted out of line and are reached
only once the static branch is patched on:
endbr64 ; 4 bytes
xchg %ax,%ax ; 2 static-branch NOP
; (added)
movb $0x0,(%rdi) ; 3 unlock (single store)
A: decl %gs:__preempt_count ; 7
je B ; 2
jmp __x86_return_thunk ; 5
call queued_spin_release_traced ; 5 out of line, reached
; only when the
; tracepoint is on
jmp A ; 2 (added)
B: call __SCT__preempt_schedule ; 5
jmp __x86_return_thunk ; 5
Baseline is the same stream without the NOP and the out-of-line
CALL/JMP: 31 bytes vs 40 (+9 bytes).
Binary size impact on x86_64, defconfig: +680 bytes (+0.00%), since all
standard configs out-of-line unlock. Architectures with inlined unlock
(s390 (always), csky and loongarch (both when !PREEMPTION)) will see a
bigger increase in binary size.
On the same path (x86_64, PARAVIRT_SPINLOCKS=n) with the tracepoint
disabled, a _raw_spin_unlock()-heavy nginx workload [1] shows no
measurable difference between baseline and patched kernels in
throughput, latency, cycles, instructions, IPC, or L1 instruction-cache
misses (kernel and total): all deltas stay within run-to-run noise.
Unlike x86, on arm64 the frame setup code (STP, MOV and LDP) lands on
the executed path in addition to static-branch NOP. Binary size impact
on arm64, defconfig: +932 bytes (+0.00%).
The _raw_spin_unlock()-heavy nginx workload reflects the larger hot
path: L1 instruction-cache misses rise ~1.4% (kernel and total) and
instruction count ~0.4%, consistent with the per-unlock frame.
cpu_cycles, throughput and latency show no measurable change and are
within run-to-run noise.
Architectures with fully custom qspinlock implementations (e.g.
PowerPC) are not covered by this change.
[1]: https://lore.kernel.org/all/aiphFXe_TPNPxZ_n@shell.ilvokhin.com/
Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juergen Gross <jgross@suse.com>
Link: https://patch.msgid.link/0d998e22a0c595f670cfc6725bb683323aced5cb.1785778551.git.d@ilvokhin.com
|
|
The contended_release tracepoint needs to hook queued_spin_unlock(), but
architectures with a custom unlock define queued_spin_unlock() directly,
leaving no single generic place to add the tracing.
Introduce queued_spin_release() as the arch-overridable release
primitive and make queued_spin_unlock() a generic wrapper around it.
An architecture that only customizes the release can then override
queued_spin_release() and inherit the generic wrapper.
Rename the MIPS override to queued_spin_release() accordingly. x86
paravirt overrides queued_spin_unlock() directly and is left unchanged.
No functional change intended.
Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juergen Gross <jgross@suse.com>
Link: https://patch.msgid.link/b8daabae6469ad72cc784a911f6cc43a6d45df3a.1785778551.git.d@ilvokhin.com
|
|
Merge ACPI bus type code updates, ACPI sysfs code updates related to
exposing table data, updates of ACPI data-only table parsers and
assorted driver changes related to ACPI support for 7.3-rc1:
- Eliminate struct acpi_driver whose users have all been converted to
bind to platform devices or auxiliary devices and set the "no power
management" flag for all struct acpi_device objects (Rafael Wysocki)
- Update kerneldoc comments of two structures in the ACPI bus type code
to use correct struct member names to avoid warnings (Randy Dunlap)
- Avoid complaints regarding missing _OSC features on platforms where
OSC_CAPABILITIES_MASK_ERROR is set in _OSC error bits even though all
of the requested features are actually acknowledged (Rafael Wysocki)
- Use correct region struct for BERT region size check and properly
map BERT and CCEL data to their ACPI tables (Thomas Renninger)
- Clean up the list of included header files in the NHLT table parser
and validate the table and record lengths in the FPDT parser (Andy
Shevchenko and Pengpeng Hou)
- Stop using acpi_device_name() in the PNP core, stop setting
acpi_device_name/class() in the Xen variant of the ACPI PAD
(Processor Aggregator Device) driver, and make the Loongarch
laptop driver stop setting acpi_device_class() (Rafael Wysocki)
* acpi-bus:
ACPI: bus: Avoid confusing complaints regarding missing _OSC features
ACPI: bus: Use correct struct member names
ACPI: scan: Set power.no_pm for all struct acpi_device objects
ACPI: bus: Eliminate struct acpi_driver
* acpi-sysfs:
ACPI: sysfs: Properly map BERT and CCEL data to their ACPI tables
ACPI: Use correct region struct for BERT region size check
* acpi-tables:
ACPI: NHLT: Remove always included kconfig.h
ACPI: FPDT: validate table and record lengths
* acpi-driver:
PNP: ACPI: Stop using acpi_device_name()
ACPI: PAD: xen: Stop setting acpi_device_name/class()
platform/loongarch: laptop: Stop setting acpi_device_class()
|
|
Hyeoncheol Jeong <hyenc.jeong@samsung.com> says:
UFS 5.0 / JEDEC 220H introduces the AGGREGATED READ query opcode
(0x9), which retrieves an aggregated data packet bundling multiple
Descriptors, Attributes and Flags in a single QUERY RESPONSE
UPIU. Such a packet can be far larger than a single descriptor, so the
reserved (device management) tag gets a dedicated UCD with an enlarged
response area.
Patch 1 is a preparatory cleanup that switches the BSG query
descriptor length to unsigned types. Patch 2 adds the aggregated read
support.
v3: https://lore.kernel.org/linux-scsi/20260724030812epcms2p4eb2c77cb4dcd5fd51ca9c3eaa5ea4bfa@epcms2p4/
v2: https://lore.kernel.org/linux-scsi/20260722084819epcms2p49c27fce999e821385f7b5d7ea5a02868@epcms2p4/
Link: https://patch.msgid.link/20260728092208epcms2p881b17276fb41c006a4229c1d073b4ad0@epcms2p8
Signed-off-by: Martin K. Petersen (Oracle) <mkp@kernel.org>
|
