| Age | Commit message (Collapse) | Author |
|---|
|
When performing data relocation on a zoned filesystem, BTRFS can deadlock
in handle_reserve_tickets(). The relocation process is waiting on a space
reservation ticket that can never be fulfilled, because the relocation
itself is the operation responsible for freeing up that space.
Fix this by introducing a new flush state,
BTRFS_RESERVE_FLUSH_ZONED_RELOCATION, specifically for data chunk
allocation during zoned relocation. Like
BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE, this state uses
priority_reclaim_data_space() instead of the normal flushing path, which
avoids re-entering the relocation code and breaking the deadlock cycle.
In btrfs_alloc_data_chunk_ondemand(), select this new flush state when the
inode belongs to a data relocation root on a zoned filesystem.
Fixes: e2a7fd22378f ("btrfs: zoned: add zone reclaim flush state for DATA space_info")
Reviewed-by: Boris Burkov <boris@bur.io>
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Don't account the free space in a data relocation space-info sub-group as
usable free space in statfs.
This is misleading as no user allocations can be made in this space-info
sub-group. It is only a target for relocation.
Fixes: f92ee31e031c ("btrfs: introduce btrfs_space_info sub-group")
Reviewed-by: Boris Burkov <boris@bur.io>
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
When searching for a data relocation block-group on mount,
btrfs_zoned_reserve_data_reloc_bg() is looking for the first empty DATA
block-group. But it first checks if the block-group is empty and if yes
continues the search, and then checks if it is the first DATA block-group.
There is actually no point in looking for the second empty DATA block
group as new DATA allocations will just allocate a new chunk for it. Pick
the first DATA block-group without any allocations done and set it as
relocation block-group.
At first, the commit 694ce5e143d6 ("btrfs: zoned: reserve data_reloc
block group on mount") introduced the functionality. At that time, we
took second unused (used == 0) block group, as the first one might be a
block group used for normal data. Later, commit daa0fde32235 ("btrfs:
zoned: fix data relocation block group reservation") switched to look
for an empty block group (alloc_offset == 0). At this point, there is no
reason taking the second one anymore. So, this commit is fixing an issue
in commit daa0fde32235.
Reviewed-by: Boris Burkov <boris@bur.io>
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Add support for ASPEED SoC family like ast27XX 8th
generation ASPEED BMCs.
Signed-off-by: Ryan Chen <ryan_chen@aspeedtech.com>
Link: https://lore.kernel.org/r/20260609-upstream_ast2700-v9-2-f631752f0cb1@aspeedtech.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Add device tree compatible string for AST2700 based boards
("aspeed,ast2700-evb" and "aspeed,ast2700") to the Aspeed SoC
board bindings. This allows proper schema validation and
enables support for AST2700 platforms.
Signed-off-by: Ryan Chen <ryan_chen@aspeedtech.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20260609-upstream_ast2700-v9-1-f631752f0cb1@aspeedtech.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/conor/linux into soc/dt
Microchip RISC-V devicetrees for v7.2
This time around, there's nothing other than patches for Microchip
boards. All of this is low priority fixes and cleanup, centred on the
pic64gx and beaglev-fire boards. There is no new support added.
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
* tag 'riscv-dt-for-v7.2' of https://git.kernel.org/pub/scm/linux/kernel/git/conor/linux:
riscv: dts: microchip: remove redudant enabling of syscontroller
riscv: dts: microchip: fix pic64gx gpio interrupt-cells
riscv: dts: microchip: add gpio line names on beaglev-fire
riscv: dts: microchip: add adc interrupt on beaglev-fire
riscv: dts: microchip: clean up beaglev-fire regulator node names
riscv: dts: microchip: remove gpio hogs from beaglev-fire
riscv: dts: microchip: gpio controllers on mpfs need 2 interrupt cells
riscv: dts: microchip: sort pic64gx i2c nodes alphanumerically
riscv: dts: microchip: update pic64gx gpio interrupts to better match the SoC
riscv: dts: microchip: add tsu clock to macb on pic64gx
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Increase the pm_runtime autosuspend delay to be longer than the
timeout of the firmware's own inactivity timer.
There is no point attempting to pm_runtime suspend any sooner than
the firmware idle timeout because it would only mean the driver has
to poll waiting for the firmware idle.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://patch.msgid.link/20260609122946.288103-1-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Now when I/O is submitted to a partition, the per-CPU in_flight[]
counter is incremented only on the partition's block_device, not on the
underlying whole disk. This leads to a problem which can be shown by a
fio test:
lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
mydev 252:1 0 20G 0 disk
└─mydev1 259:0 0 10G 0 part
iostat -xp 1
Device r/s rkB/s ... aqu-sz %util
mydev 128153.00 512612.00 ... 13.22 72.20
mydev1 128154.00 512616.00 ... 13.22 100.00
%util is different between mydev and mydev1, which is unexpected.
This is the cumulative effect of a series of patches. The root cause is
commit e016b78201a2 ("block: return just one value from part_in_flight"),
which deleted the branch in part_in_flight() that aggregated the whole-disk
in_flight count on top of the partition's. Then the second commit is
commit 10ec5e86f9b8 ("block: merge part_{inc,dev}_in_flight into their
only callers"), which folded the whole-disk in_flight accounting into
generic_start_io_acct() and generic_end_io_acct(). Those two helpers
were then removed by commit e722fff238bb ("block: remove
generic_{start,end}_io_acct"), and from that point on the whole disk's
in_flight is no longer accounted at all.
In update_io_ticks(), if calling bdev_count_inflight() finds that the
inflight value of the whole device is 0, the accumulation of io_ticks will
be skipped, causing the reported util% value to be underestimated.
Fix it by restoring the whole-disk in_flight accounting.
Fixes: e016b78201a2 ("block: return just one value from part_in_flight")
Suggested-by: Leon Hwang <leon.huangfu@shopee.com>
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Tang Yizhou <yizhou.tang@shopee.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://patch.msgid.link/20260526021555.359500-1-yizhou.tang@shopee.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into soc/dt
We've got basic camera support on RK3588!
New peripherals RK3588 vicap plus camera addition to some boards,
RK3528 USB and USB enablement on some boards, RGA3 support on RK3588.
Missing EL2 virtual timer interrupt added to RK3588.
Some more added peripherals for the Khadas Edge 2L board.
* tag 'v7.2-rockchip-dts64-2' of https://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:
arm64: dts: rockchip: Fix vcc_sdio regulator max voltage on Pinebook Pro
arm64: dts: rockchip: Enable USB 2.0 ports on NanoPi Zero2
arm64: dts: rockchip: Enable USB 2.0 ports on ArmSoM Sige1
arm64: dts: rockchip: Enable USB ports on Radxa ROCK 2A/2F
arm64: dts: rockchip: Enable USB 2.0 ports on Radxa E20C
arm64: dts: rockchip: Add USB nodes for RK3528
arm64: dts: rockchip: enable adc button for Radxa E25
arm64: dts: rockchip: Add Bluetooth support for Khadas Edge 2L
arm64: dts: rockchip: Enable USB for Khadas Edge 2L
arm64: dts: rockchip: Disable removed devices from rk3399-nanopi-r4s
arm64: dts: rockchip: Fix EEPROM compatible on rk3399-nanopi-r4s-enterprise
arm64: dts: rockchip: add radxa camera 4k on rock 5b+ cam1
arm64: dts: rockchip: add radxa camera 4k on rock 5b+ cam0
arm64: dts: rockchip: add vicap node to rk3588
arm64: dts: rockchip: Add EL2 virtual timer interrupt
arm64: dts: rockchip: add rga3 dt nodes to rk3588
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
cpu_latency_qos_init() registers cpu_dma_latency first and, when
CONFIG_PM_QOS_CPU_SYSTEM_WAKEUP is enabled, registers cpu_wakeup_latency
afterwards. The second registration overwrites the first return value.
As a result, a failure to register cpu_dma_latency can be masked if the
second registration succeeds. Conversely, if cpu_dma_latency succeeds and
cpu_wakeup_latency fails, the function returns an error while leaving the
first misc device registered.
Return immediately on the first registration failure and deregister
cpu_dma_latency if the second registration fails.
Fixes: a4e6512a79d8 ("PM: QoS: Introduce a CPU system wakeup QoS limit")
Signed-off-by: Yuho Choi <dbgh9129@gmail.com>
Link: https://patch.msgid.link/20260608170748.82273-1-dbgh9129@gmail.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
|
|
virtblk_report_zones() trusts the device-reported number of zones when
walking the report buffer:
nz = min_t(u64, virtio64_to_cpu(vblk->vdev, report->nr_zones),
nr_zones);
...
for (i = 0; i < nz && zone_idx < nr_zones; i++) {
ret = virtblk_parse_zone(vblk, &report->zones[i], ...);
The buffer is allocated by virtblk_alloc_report_buffer(), whose size is
capped by the queue's max hardware sectors and max segments and can
therefore hold fewer descriptors than nr_zones. nz is bounded only by
the device-supplied report->nr_zones and the requested nr_zones, never
by the buffer's descriptor capacity. At probe time the request count is
unbounded (blk_revalidate_disk_zones() calls report_zones() with
nr_zones == UINT_MAX), so the device-supplied report->nr_zones is the
sole gate: a device that reports more zones than fit in the buffer
drives the loop to read report->zones[i] past the end of the allocation.
A malicious or buggy virtio-blk device that reports an inflated nr_zones
triggers this during zone revalidation at probe. KASAN reports a
vmalloc-out-of-bounds read in virtblk_report_zones() against the report
buffer allocated a few lines earlier.
Clamp nz to the number of descriptors that actually fit in the report
buffer.
Fixes: 95bfec41bd3d ("virtio-blk: add support for zoned block devices")
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://patch.msgid.link/20260607124834.3059944-1-michael.bommarito@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
into soc/dt
ARM64: DT: HiSilicon ARM64 DT updates for v7.2
- Move role-switch endpoint into connector on hi3660-hikey960
* tag 'hisi-arm64-dt-for-7.2' of https://github.com/hisilicon/linux-hisi:
arm64: dts: hisilicon: hi3660-hikey960: move role-switch endpoint into connector
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/frank.li/linux into soc/dt
i.MX ARM device tree changes for 7.2:
DT Binding Cleanup:
- Replaced undocumented compatible strings with proper ones:
* edt,edt-ft5x06 -> edt,edt-ft5206
* marvell,88E1510 -> ethernet-phy-ieee802.3-c22
* karo,imx6qdl-tx6-sgtl5000 -> simple-audio-card
- Fixed incorrect VAR-SOM-MX6UL references (corrected to VAR-SOM-MX6)
- Added missing required properties:
* #phy-cells for usb-nop-xceiv
* #io-channel-cells to ADC nodes
* bus-type for ov5642/ov5640 cameras
* ti,deskew = <0> for ti,tfp410
- Added missing supply properties (power-supply, vdd-supply, dvdd-supply, avdd-supply)
- Removed redundant/empty properties (bus-width for video-mux, empty clock-names)
- Fixed boolean property warnings and non-existent property references
- Converted TS-4800 watchdog to DT schema
- Renamed wdt nodes to watchdog for consistency
New Features Added:
- PCIe Root Port nodes and PERST property for imx6qdl, imx6sx, and imx7d
- OV5645 camera support for imx7d-pico-pi
- LVDS display panel support for imx6ul-var-som
- WiFi and Bluetooth support for VAR-SOM boards
- nvmem-layout support for imx7
- New bus bindings: fsl,aipi-bus and fsl,emi-bus
- New board binding: variscite,var-som-imx6ull
Code Refactoring:
- VAR-SOM-MX6UL/ULL: factored out common parts for CPU variants
- Separated audio, ethernet (ENET1/ENET2), and SD card support into reusable components
* tag 'imx-dt-7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/frank.li/linux: (35 commits)
dt-bindings: soc: imx: Add fsl,aipi-bus and fsl,emi-bus
ARM: dts: freescale: add bootph-all to i.MX7ULP watchdog nodes
ARM: dts: imx7: add nvmem-layout
ARM: dts: imx7d-pico-pi: add OV5645 camera support
ARM: dts: imx6-display5: replace marvell,88E1510 with ethernet-phy-ieee802.3-c22
ARM: dts: imx: replace undocumented compatible string edt,edt-ft5x06 with edt,edt-ft5206
ARM: dts: imx6qdl-tx6: remove undocumented karo,imx6qdl-tx6-sgtl5000 and keep only simple-audio-card
ARM: dts: imx: Add bus-type for ov5642/ov5640
ARM: dts: imx: remove redundant bus-width for video-mux
ARM: dts: imx: add (power|vdd)-supply for related node
ARM: dts: imx53-ppd: add '#phy-cells' for usb-nop-xceiv
ARM: dts: imx53-qsb: add dvdd and avdd supply for panel sii,43wvf1g
ARM: dts: imx: add ti,deskew = <0> for ti,tfp410
ARM: dts: imx7d: Add Root Port node and PERST property
ARM: dts: imx6sx: Add Root Port node and PERST property
ARM: dts: imx6qdl: Add Root Port node and PERST property
ARM: dts: nxp: imx51-ts4800: Rename wdt node to watchdog
dt-bindings: watchdog: Convert TS-4800 to DT schema
ARM: dts: imx6ul: add #io-channel-cells to ADC
ARM: dts: imx25: remove empty clock-names for nand-controller@bb000000
...
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/sven/linux into soc/dt
Apple SoC DT update for 7.2
Add minimal device trees for all t8122 / base M3 based devices and some
required new compatibles to the dt-bindings. These are enough to boot
Linux on these devices to a simple serial console but future work is
required to make these machines useful for end users.
Signed-off-by: Sven Peter <sven@kernel.org>
* tag 'apple-soc-dt-7.2' of https://git.kernel.org/pub/scm/linux/kernel/git/sven/linux:
arm64: dts: apple: Initial t8122 (M3) device trees
dt-bindings: arm: apple: Add M3 based devices
dt-bindings: pwm: apple,s5l-fpwm: Add t8122 compatible
dt-bindings: power: apple,pmgr-pwrstate: Add t8122 compatible
dt-bindings: arm: apple: apple,pmgr: Add t8122 compatible
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/ti/linux into soc/dt
TI K3 device tree updates for v7.2
SoC Specific Features and Fixes:
J722S:
- Use ti,j7200-padconf compatible for pad configuration
- Add MCU and wakeup domain peripherals specific to J722S
- Use J722S-specific compatibles for WIZ, gmii-sel and CPSW3G nodes
Board Specific Features and Fixes:
AM62x (Toradex Verdin):
- Add display overlays: DSI-to-HDMI adapter, DSI-to-LVDS adapter with 10.1"
panel, capacitive touch displays in 7" DSI, 10.1" DSI and 10.1" LVDS
configurations, and Mezzanine with 10.1" LVDS display
- Add NAU8822 Bridge Tied Load audio support
- Add OV5640 CSI camera overlays
- Add Verdin Mezzanine CAN overlay
- Reserve UART_4 for Cortex-M4F use
AM625:
- Add support for TQ-Systems TQMa62xx SoM and MBa62xx carrier board,
including new dt-bindings compatible strings
AM62x (phyBOARD-Lyra):
- Add DT overlay for Lincoln LCD185-101CT OLDI panel
AM62-LP SK:
- Add system-power-controller node
AM62A7 SK:
- Add bootph-all tag to vqmmc regulator for proper boot sequencing
J721S2-som:
- Add bootph-pre-ram property to PMIC-B for proper early boot sequencing
* tag 'ti-k3-dt-for-v7.2' of https://git.kernel.org/pub/scm/linux/kernel/git/ti/linux:
arm64: dts: ti: k3-am62-verdin: Add Mezzanine with Toradex Display 10.1" LVDS
arm64: dts: ti: k3-am62-verdin: Add Toradex Verdin Mezzanine CAN
arm64: dts: ti: k3-am62-verdin: Add Toradex OV5640 CSI Cameras
arm64: dts: ti: k3-am62-verdin: Reserve UART_4 for Cortex-M4F
arm64: dts: ti: k3-am62-verdin: Add NAU8822 Bridge Tied Load
arm64: dts: ti: k3-am62-verdin: Add Toradex Capacitive Touch Display 7" DSI
arm64: dts: ti: k3-am62-verdin: Add Toradex Capacitive Touch Display 10.1" DSI
arm64: dts: ti: k3-am62-verdin: Add Toradex Capacitive Touch Display 10.1" LVDS
arm64: dts: ti: k3-am62-verdin: Add Toradex DSI to LVDS adapter with 10.1" display
arm64: dts: ti: k3-j722s-main: use J722S compatibles for WIZ, gmii-sel and CPSW3G
arm64: dts: ti: k3-am62-verdin: Add DSI to HDMI adapter overlay
arm64: dts: ti: Add TQ-Systems TQMa62xx SoM and MBa62xx carrier board Device Trees
dt-bindings: arm: ti: Add compatible for AM625-based TQMa62xx SOM family and carrier board
arm64: dts: ti: am62-phyboard-lyra: Add DT overlay for Lincoln LCD185-101CT panel
arm64: dts: ti: k3-j721s2-som-p0: add bootph-pre-ram property to PMIC-B
arm64: dts: ti: k3-j722s: Add wakeup domain peripherals specific to J722S
arm64: dts: ti: k3-j722s: Add mcu domain peripherals specific to J722S
arm64: dts: ti: k3-j722s: Use ti,j7200-padconf compatible
arm64: dts: ti: k3-am62-lp-sk: Add system-power-controller
arm64: dts: ti: k3-am62a7-sk: Add bootph-all tag to vqmmc
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
https://github.com/Broadcom/stblinux into soc/dt
This pull request contains Broadcom ARM-based SoCs Device Tree changes
for 7.2, please pull the following:
- Rosen moves the Meraki MX6X pinctrl configuration to the PWM node
where it belongs, also fixes the USB3 GPIO for the R6300v2 and
EA6500v2 routers
- Jinseok fixes a gpio label for the bcm2711 systems (Raspberry Pi 4)
* tag 'arm-soc/for-7.2/devicetree' of https://github.com/Broadcom/stblinux:
arm: dts: bcm2711: Fix typo in gpio-line-names
ARM: dts: BCM5301X: EA6500v2: fix USB3
ARM: dts: BCM5301X: R6300v2: fix USB3
ARM: dts: NSP: Move MX6X pinctrl config to PWM node
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux into soc/dt
arm64: tegra: Changes for v7.2-rc1
This contains a fix for GPIO on Jetson AGX Thor and enables the PCIe
controllers used on this platform.
Smaug (a.k.a. Pixel C) devices can now be properly powered off via the
PMIC.
Miscellaneous improvements are made across a number of different boards,
such as SMMU enablement on Tegra194 platforms and some better bootloader
interoperation on Chromium-based devices.
Finally, various minor fixes to device tree properties round off this
set of changes.
* tag 'tegra-for-7.2-arm64-dt' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
arm64: tegra: Enable SMMU on Tegra194 display controllers
Revert "arm64: tegra: Disable ISO SMMU for Tegra194"
arm64: tegra: Add #{address,size}-cells to Chromium-based /firmware
arm64: tegra: Fix aspm-l1-entry-delay-ns L1 latency cells
arm64: tegra: Mark MAX77620 as system power controller on Smaug
arm64: tegra: Enable PCIe for Jetson AGX Thor
arm64: tegra: Fix address of Tegra264 main GPIO controller
arm64: tegra: Add aspm-l1-entry-delay-ns to PCIe nodes
arm64: tegra: Fix Tegra234 MGBE PTP clock
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
According to the PHY Databook Common Block Signals section, the
ref_clk_en signal must remain de-asserted until the reference clock is
running at the appropriate frequency. Once the clock is stable,
ref_clk_en can be asserted. For lower power states where the reference
clock to the PHY is disabled, ref_clk_en should also be de-asserted.
Move the ref_clk_en bit manipulation into imx95_pcie_enable_ref_clk()
to ensure the reference clock stabilizes before ref_clk_en is asserted
and before the PHY reset is de-asserted. This aligns with the timing
requirements specified in the PHY documentation.
Fixes: d8574ce57d76 ("PCI: imx6: Add external reference clock input mode support")
Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
Signed-off-by: Manivannan Sadhasivam <mani@kernel.org>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260518072715.3166514-3-hongxing.zhu@nxp.com
|
|
According to the i.MX95 PCIe PHY Databook, the ref_use_pad signal in the
Common Block Signals section selects the reference clock source connected
to the PHY pads. Per the specification, any change to this input must be
followed by a PHY reset assertion to take effect.
Move the REF_USE_PAD configuration before the PHY reset toggle to comply
with the required initialization sequence.
Fixes: 47f54a902dcd ("PCI: imx6: Toggle the core reset for i.MX95 PCIe")
Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
[mani: renamed the callback and helper to match the usecase]
Signed-off-by: Manivannan Sadhasivam <mani@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260518072715.3166514-2-hongxing.zhu@nxp.com
|
|
Nuoqi Gui says:
====================
bpf: Enforce BTF pointer write checks for global args
check_mem_reg() verifies both read and write access when a caller passes
memory into a global subprogram. For PTR_TO_BTF_ID callers,
check_helper_mem_access() currently always checks the access as BPF_READ.
That lets a tracing program pass a task_struct field pointer to a global
subprogram argument typed as writable memory. The direct field store is rejected
with "only read is supported", but the callee is validated with a generic
writable PTR_TO_MEM argument and can store through it.
Forward the requested access type into the PTR_TO_BTF_ID helper-access path and
add verifier coverage for the global-subprogram argument case.
Validation (tested on bpf-next 8496d9020ff3):
Without this series:
direct BTF field store rejected with "only read is supported";
global-subprogram candidate loaded, attached, and runtime-confirmed.
With this series applied:
direct BTF field store rejected with "only read is supported";
global-subprogram candidate rejected with "only read is supported".
Signed-off-by: Nuoqi Gui <gnq25@mails.tsinghua.edu.cn>
---
====================
Link: https://patch.msgid.link/20260609-f01-04-btf-writable-arg-v1-0-f449cd970669@mails.tsinghua.edu.cn
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
|
|
Add a verifier test for passing a BTF-backed task_struct field pointer to a
global subprogram argument typed as writable memory.
The direct field store is already rejected.
The global subprogram path should be rejected too.
The callee must not lose the BTF pointer's read-only provenance.
It must not validate the argument as ordinary writable memory.
Signed-off-by: Nuoqi Gui <gnq25@mails.tsinghua.edu.cn>
Link: https://lore.kernel.org/bpf/20260609-f01-04-btf-writable-arg-v1-2-f449cd970669@mails.tsinghua.edu.cn
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
|
|
check_mem_reg() verifies both read and write access for global subprogram
memory arguments. When the caller register is PTR_TO_BTF_ID,
check_helper_mem_access() currently forwards the access to
check_ptr_to_btf_access() as BPF_READ regardless of the requested access
type.
This lets a BTF-backed kernel object field pointer pass the caller-side
writable memory check for a global subprogram argument. The callee is then
validated with a generic writable PTR_TO_MEM argument and can store through
it, even though an equivalent direct BTF field store is rejected with "only
read is supported".
Forward the requested access type to check_ptr_to_btf_access().
This enforces existing BTF write restrictions for global subprogram memory
arguments as well.
Fixes: 3e30be4288b3 ("bpf: Allow helpers access trusted PTR_TO_BTF_ID.")
Signed-off-by: Nuoqi Gui <gnq25@mails.tsinghua.edu.cn>
Link: https://lore.kernel.org/bpf/20260609-f01-04-btf-writable-arg-v1-1-f449cd970669@mails.tsinghua.edu.cn
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
|
|
Document the KVM_CAP_S390_HPAGE_2G capability, which behaves very
similarly to the existing KVM_CAP_S390_HPAGE_1M.
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260609150930.665370-5-imbrenda@linux.ibm.com>
|
|
Change gmap_2g_allowed() to perform the necessary checks to allow for
2G hugepages to be used, instead of returning false. The
GMAP_FLAG_ALLOW_HPAGE_2G gmap flag is now taken into account.
Also add appropriate kerneldoc comments.
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260609150930.665370-4-imbrenda@linux.ibm.com>
|
|
Add KVM_CAP_S390_HPAGE_2G to signal to userspace that 2G hugepages may
be used to back the guest; restrictions apply similar to 1M hugepages.
Enable the (for now still ignored) GMAP_FLAG_ALLOW_HPAGE_2G flag for
the guest gmap, and propagate / disable it as necessary.
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260609150930.665370-3-imbrenda@linux.ibm.com>
|
|
Add the hpage_2g module parameter to KVM to allow enabling or disabling
2G hugepages in KVM.
If hpage_2g is enabled but hpage is not enabled, print a message and
disable hpage_2g.
Opportunistically fix the comment for the hpage module parameter.
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260609150930.665370-2-imbrenda@linux.ibm.com>
|
|
The LTC4283 device has up to 8 pins that can be configured as GPIOs.
Note that PGIO pins are not set as GPIOs by default so if they are
configured to be used as GPIOs we need to make sure to initialize them
to a sane default. They are set as inputs by default.
Acked-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Reviewed-by: Linus Walleij <linusw@kernel.org>
Signed-off-by: Nuno Sá <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20260502-ltc4283-support-v13-3-1c206542e652@analog.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Support the LTC4283 Hot Swap Controller. The device features programmable
current limit with foldback and independently adjustable inrush current to
optimize the MOSFET safe operating area (SOA). The SOA timer limits MOSFET
temperature rise for reliable protection against overstresses.
An I2C interface and onboard ADC allow monitoring of board current,
voltage, power, energy, and fault status.
Signed-off-by: Nuno Sá <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20260502-ltc4283-support-v13-2-1c206542e652@analog.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Some workloads with different CPU capacities consume more power with
timer migration than before. The recently introduced per capacity
hierarchies were supposed to alleviate this problem. However it appears
to also regress other types of workloads, especially when plenty of
capacities live together in the same machine.
Disable the feature until a reasonable solution is found.
Fixes: 098cbaad8e57 ("timers/migration: Split per-capacity hierarchies")
Reported-by: Christian Loehle <christian.loehle@arm.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260609123356.28449-1-frederic@kernel.org
Closes: https://lore.kernel.org/all/3b79338f-6cfc-4722-8062-9103db2c8ad1@arm.com
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull misc fixes from Andrew Morton:
"11 hotfixes. 9 are for MM. 8 are cc:stable and the remaining 3 address
post-7.1 issues or aren't considered suitable for backporting.
Thre's a two-patch series "mm/damon/{reclaim,lru_sort}: handle ctx
allocation failures" from SeongJae Park which fixes a couple of DAMON
-ENOMEM bloopers. The rest are singletons - please see the individual
changelogs for details"
* tag 'mm-hotfixes-stable-2026-06-08-20-51' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
mm/mincore: handle non-swap entries before !CONFIG_SWAP guard
arm64: mm: call pagetable dtor when freeing hot-removed page tables
mm/list_lru: drain before clearing xarray entry on reparent
mm/huge_memory: use correct flags for device private PMD entry
mm/damon/lru_sort: handle ctx allocation failure
mm/damon/reclaim: handle ctx allocation failure
zram: fix use-after-free in zram_bvec_write_partial()
MAINTAINERS: update Baoquan He's email address
tools headers UAPI: sync linux/taskstats.h for procacct.c
mm/cma_sysfs: skip inactive CMA areas in sysfs
ipc/shm: serialize orphan cleanup with shm_nattch updates
|
|
The LTC4283 is a negative voltage hot swap controller that drives an
external N-channel MOSFET to allow a board to be safely inserted and
removed from a live backplane.
Special note for the "adi,vpower-drns-enable" property. It allows to choose
between the attenuated MOSFET drain voltage or the attenuated input
voltage at the RTNS pin (effectively choosing between input or output
power). This is a system level decision not really intended to change at
runtime and hence is being added as a Firmware property.
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Nuno Sá <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20260502-ltc4283-support-v13-1-1c206542e652@analog.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Fix driver issues:
- Add the missing regulator and property files in include
- Declare XDP720_DEFAULT_RIMON as unsigned constant
- Declare struct pmbus_driver_info xdp720_info as constant
Signed-off-by: Ashish Yadav <ashish.yadav@infineon.com>
Link: https://lore.kernel.org/r/20260609072231.15486-4-Ashish.Yadav@infineon.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Adds support for the Infineon XDP730 Digital eFuse Controller by
updating the existing XDP720 driver.
Signed-off-by: Ashish Yadav <ashish.yadav@infineon.com>
Link: https://lore.kernel.org/r/20260609072231.15486-3-Ashish.Yadav@infineon.com
[groeck: Fixed conflicts in xdp720_id declaration]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add documentation for the device tree binding of the XDP730 eFuse.
Rename node to efuse to accurately reflect its hardware function.
Signed-off-by: Ashish Yadav <ashish.yadav@infineon.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20260609072231.15486-2-Ashish.Yadav@infineon.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add of_match_table to add support of devicetree probing.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
[rgantois: Removed of_match_ptr().]
Signed-off-by: Romain Gantois <romain.gantois@bootlin.com>
Link: https://lore.kernel.org/r/20260608-adt7462-bindings-v2-1-272982c40325@bootlin.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add support for ROG MAXIMUS Z790 EXTREME
Signed-off-by: Brian Downey <bdowne01@gmail.com>
Signed-off-by: Eugene Shalygin <eugene.shalygin@gmail.com>
Link: https://lore.kernel.org/r/20260608060855.40469-1-eugene.shalygin@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add a PMBus driver for the Analog Devices MAX20860A step-down DC-DC
switching regulator. The MAX20860A provides monitoring of input/output
voltage, output current, and temperature via the PMBus interface using
linear data format. Optional regulator support is available via
CONFIG_SENSORS_MAX20860A_REGULATOR.
Signed-off-by: Syed Arif <arif.syed@hpe.com>
Signed-off-by: Sanman Pradhan <psanman@juniper.net>
Link: https://lore.kernel.org/r/20260601184516.919488-3-sanman.pradhan@hpe.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add devicetree binding documentation for the Analog Devices MAX20860A
step-down DC-DC switching regulator with PMBus interface.
Signed-off-by: Sanman Pradhan <psanman@juniper.net>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20260601184516.919488-2-sanman.pradhan@hpe.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add support for BMR316, BMR321, BMR350 and BMR351 DC/DC converter
modules from Flex to the pmbus driver.
Signed-off-by: Daniel Nilsson <linux@erq.se>
Link: https://lore.kernel.org/r/20260603085712.659432-2-linux@erq.se
[groeck: Resolved conflicts (explicit struct members in pmbus_id)]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Named initializers are better readable and more robust to changes of the
struct definition. This robustness is relevant for a planned change to
struct platform_device_id replacing .driver_data by an anonymous unit.
While touching these arrays unify usage of commas.
Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@baylibre.com>
Link: https://lore.kernel.org/r/25d38df8db42d69f33fa30267c9fd5ea058223d0.1779894738.git.u.kleine-koenig@baylibre.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
The driver explicitly set the .driver_data member of struct
platform_device_id to zero without relying on that value. Drop this
unused assignments.
While touching this array unify spacing and use named initializers for
.name.
Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@baylibre.com>
Reviewed-by: Tzung-Bi Shih <tzungbi@kernel.org>
Acked-by: Thomas Weißschuh <linux@weissschuh.net>
Link: https://lore.kernel.org/r/972c9998054c7944f63266819d6fb08b36edb5c5.1779894738.git.u.kleine-koenig@baylibre.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
set_fan() parses user input with kstrtol() and passes the resulting
value to FAN16_TO_REG() on chips with 16-bit fan support.
Negative fan speeds are not meaningful and should be rejected before
conversion. Worst scenario, one may be able to abuse undefined
behaviour of signed overflow to possibly induce rpm * 2 == 0 in
FAN16_TO_REG(), thus causing a division by zero.
Instead, clamp val < 0 to zero and keep the conversion in its valid
input domain, avoiding unsafe arithmetic in the register conversion
path.
Found by Linux Verification Center (linuxtesting.org) with static
analysis tool SVACE.
Fixes: 17d648bf5786 ("it87: Add support for the IT8716F")
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Link: https://lore.kernel.org/r/20260529141839.1639287-1-n.zhandarovich@fintech.ru
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
The board has a similar sensor configuration to the
ROG STRIX B850-I GAMING WIFI, but includes an additional
T-Sensor header. The patch was provided via GitHub [1].
[1] https://github.com/zeule/asus-ec-sensors/pull/105
Signed-off-by: Eugene Shalygin <eugene.shalygin@gmail.com>
Link: https://lore.kernel.org/r/20260607123626.100630-1-eugene.shalygin@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add support for ROG STRIX B650E-E GAMING WIFI
Signed-off-by: Veronika Kossmann <nanodesuu@gmail.com>
Co-developed-by: Oleg Tsvetkov <oleg-tsv@yandex.ru>
Signed-off-by: Oleg Tsvetkov <oleg-tsv@yandex.ru>
Signed-off-by: Eugene Shalygin <eugene.shalygin@gmail.com>
Link: https://lore.kernel.org/r/20260607110702.84599-2-eugene.shalygin@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Add the ASRock Z890 Pro-A customer ID to the list of supported
boards for the NCT6683 hardware monitoring driver.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Reiner Pröls <reiner.proels@gmail.com>
Link: https://lore.kernel.org/r/20260521212632.223724-1-Reiner.Proels@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Since device_property_read_string() and similar functions defined in
linux/property.h are used in the driver add explicit include for
linux/mod_devicetable.h and linux/property.h rather than having implicit
inclusions.
Removed of_match_ptr() improving non-Device Tree compatibility of the
driver and drop unnecessary __maybe_unused.
Header linux/of.h can't be removed yet since macro is_of_node() is used.
Signed-off-by: Flaviu Nistor <flaviu.nistor@gmail.com>
Link: https://lore.kernel.org/r/20260522052352.12139-1-flaviu.nistor@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
The driver caches the PWM frequency register and the CONFIG_FAN slow-clock
select bit, but never lets userspace pick a different output frequency.
Add a pwm1_freq sysfs attribute that selects the closest SCS + PFR
combination for the requested value in Hz, gated by manual mode like
set_pwm1(). PFR is clamped to 31 so that 2*PFR fits in the chip's 6-bit
PWM register (matching the existing scaling assumption in show_pwm1).
The hardware LUT hysteresis register is shared by all LUT entries, so
the per-point pwm1_auto_pointN_temp_hyst attributes can't be made RW
without N-to-1 cross-attribute side effects. Following the max31760
precedent, expose a single chip-wide pwm1_auto_point_temp_hyst attribute
holding the hysteresis amount in millidegrees; the per-point attributes
stay RO and continue to show the resulting absolute trip-down
temperature for each entry.
This was tested on a Linksys LGS328MPC switch hardware where the fan
would not spin with the default PWM Frequency, which is why this change
is required.
Signed-off-by: Jan-Henrik Bruhn <kernel@jhbruhn.de>
Link: https://lore.kernel.org/r/20260523133617.3439102-1-kernel@jhbruhn.de
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
The driver seeds the chip's SET_RTC register once at probe with
ktime_get_real_seconds(). Over a long uptime the chip's internal
seconds counter drifts away from the host's wall-clock time, so the
timestamp embedded in each blackbox record stops being meaningful
in wall-clock terms. The datasheet recommends that the host
periodically resynchronise the counter to address this; today the
driver has no userspace-facing knob for that.
Expose SET_RTC via an rtc debugfs file alongside the other adm1266
debugfs entries:
read -- returns the chip's current SET_RTC seconds counter, so
userspace can observe how far the chip has drifted from
host wall-clock without writing anything.
write -- the kernel re-reads ktime_get_real_seconds() itself and
pushes it to the chip. The write payload is ignored;
userspace does not get to supply its own timestamp
value, so there is no way for it to push a wrong time
into the chip.
A small userspace agent (chrony hook, systemd-timesyncd dispatch
script, or a periodic cron job) can write to this file to keep the
chip's counter aligned with wall-clock across long uptimes.
Both the read and write paths take pmbus_lock to serialise against
the pmbus_core's own PAGE+register sequences and against the other
adm1266 debugfs accessors that already run under the same lock.
While at it, drop the now-redundant adm1266_set_rtc() probe-time
helper. The new adm1266_rtc_set() callback does exactly the same
byte-packing and write; probe just calls adm1266_rtc_set(client, 0)
(the ignored @val argument) after pmbus_do_probe() so the
pmbus_lock acquired by the new helper has a live mutex to take.
Signed-off-by: Abdurrahman Hussain <abdurrahman@nexthop.ai>
Assisted-by: Claude-Code:claude-opus-4-7
Assisted-by: sashiko:gemini-3.1-pro-preview
Link: https://lore.kernel.org/r/20260520-adm1266-v5-3-c72ef1fac1ea@nexthop.ai
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
The ADM1266 maintains a 16-bit non-volatile POWERUP_COUNTER register
(0xE4, datasheet Rev. D, Table 93) that increments on every power
cycle and cannot be reset by the host. Each blackbox record already
embeds the counter at record time, so the standalone live value is
primarily useful for matching a captured record back to the boot it
came from when correlating logs.
Expose it as a read-only debugfs file alongside sequencer_state. The
block-read returns two payload bytes in little-endian order.
Take pmbus_lock around the block-read so the access serialises with
any pmbus_core sequence that sets PAGE on the device. Without it, a
PAGE write from another thread could interleave between a PAGE set
and a paged read elsewhere in the driver and corrupt either side's
view of the device state machine.
Signed-off-by: Abdurrahman Hussain <abdurrahman@nexthop.ai>
Assisted-by: Claude-Code:claude-opus-4-7
Assisted-by: sashiko:gemini-3.1-pro-preview
Link: https://lore.kernel.org/r/20260520-adm1266-v5-2-c72ef1fac1ea@nexthop.ai
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
The ADM1266 blackbox can be configured in two recording modes via
BLACKBOX_CONFIG[0]: cyclic, where the device overwrites the oldest
record once the 32-record buffer fills, and single, where it stops
recording until the buffer is cleared. Deployments that need to
preserve the full record history across multiple fault episodes
typically run in single mode and need a way to clear the buffer
after the records have been collected.
Expose a write-only debugfs file alongside sequencer_state. Writing
any data to it issues the documented clear-blackbox sub-command:
a 2-byte block-write to READ_BLACKBOX (0xDE) with payload
{0xFE, 0x00} (datasheet Rev. D).
The clear is taken under pmbus_lock because READ_BLACKBOX is also
used by adm1266_nvmem_read_blackbox() to walk records one at a
time; both paths run under pmbus_lock so the clear cannot
interleave mid-iteration and corrupt the read sequence.
Signed-off-by: Abdurrahman Hussain <abdurrahman@nexthop.ai>
Assisted-by: Claude-Code:claude-opus-4-7
Assisted-by: sashiko:gemini-3.1-pro-preview
Link: https://lore.kernel.org/r/20260520-adm1266-v5-1-c72ef1fac1ea@nexthop.ai
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
