Merge tag 'tegra-for-6.17-memory' of https://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux into soc/drivers
memory: tegra: Updates for v6.17-rc1
Enable support for the memory and external memory controllers found on
Tegra264.
* tag 'tegra-for-6.17-memory' of https://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
memory: tegra: Add Tegra264 MC and EMC support
dt-bindings: memory: tegra: Add Tegra264 support
Merge tag 'tegra-for-6.17-firmware' of https://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux into soc/drivers
firmware: tegra: Updates for v6.17-rc1
Add Tegra264 support for the BPMP, fix some dependency issues and clean
up some code using new OF helpers.
* tag 'tegra-for-6.17-firmware' of https://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
firmware: tegra: bpmp: Fix build failure for tegra264-only config
firmware: tegra: bpmp: Use of_reserved_mem_region_to_resource() for "memory-region"
firmware: tegra: bpmp: Add support on Tegra264
firmware: tegra: Fix IVC dependency problems
Li Jun [Wed, 4 Jun 2025 08:17:12 +0000 (16:17 +0800)]
bus: del unnecessary init var
The compiler generates initialization instructions,
which consume additional CPU cycles. the
sysc_clockdomain_init should assign a value
to 'error' before it is read.so the var don't need init
to 0.
Merge tag 'tegra-for-6.17-soc' of https://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux into soc/drivers
soc/tegra: Updates for v6.17-rc1
The bulk of this is the addition of Tegra264 support for various low-
level components. This also adds fabric descriptors for the new Tegra254
and Tegra264 chips.
* tag 'tegra-for-6.17-soc' of https://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
soc/tegra: cbb: Add support for CBB fabrics in Tegra254
soc/tegra: cbb: Add support for CBB fabrics in Tegra264
soc/tegra: cbb: Support HW lookup to get timed out target address
soc/tegra: cbb: Improve handling for per SoC fabric data
soc/tegra: cbb: Make error interrupt enable and status per SoC
soc/tegra: cbb: Change master/slave to initiator/target
soc/tegra: cbb: Clear ERR_FORCE register with ERR_STATUS
soc/tegra: Add Tegra264 APBMISC compatible string
soc/tegra: pmc: Add Tegra264 support
soc/tegra: Enable support for Tegra264
Merge tag 'mtk-soc-for-v6.17' of https://git.kernel.org/pub/scm/linux/kernel/git/mediatek/linux into soc/drivers
MediaTek soc driver updates for v6.17
This adds a single cleanup commit for the mtk-mutex driver,
clarifying the usage of the MUTEX_MOD1, MUTEX_MOD2 registers
for applying display controller sub-component mute settings
on all MediaTek SoCs.
* tag 'mtk-soc-for-v6.17' of https://git.kernel.org/pub/scm/linux/kernel/git/mediatek/linux:
soc: mediatek: mtk-mutex: Fix confusing usage of MUTEX_MOD2
Merge tag 'reset-for-v6.17' of https://git.pengutronix.de/git/pza/linux into soc/drivers
Reset controller updates for v6.17
* Support reset controllers on Kendryte K230 and SOPHGO CV1800B.
* Add RZ/V2N USB2PHY reset controller bindings
* Use auxiliary device creation helpers in reset-mpfs.
* Convert nxp,lcp1850-rgu and snps,dw-reset binding docs to DT schema.
* Enable reset-brcmstb(-rescal) on BCM2712.
* Fix a typo in the T-HEAD TH1520 Kconfig option
* tag 'reset-for-v6.17' of https://git.pengutronix.de/git/pza/linux:
dt-bindings: reset: Convert snps,dw-reset to DT schema
reset: brcmstb: Enable reset drivers for ARCH_BCM2835
reset: simple: add support for Sophgo CV1800B
dt-bindings: reset: sophgo: Add CV1800B support
reset: mpfs: use the auxiliary device creation
dt-bindings: reset: renesas,rzv2h-usb2phy: Document RZ/V2N SoC support
dt-bindings: reset: convert nxp,lpc1850-rgu.txt to yaml format
reset: thead: Fix TH1520 typo
reset: canaan: add reset driver for Kendryte K230
dt-bindings: reset: add support for canaan,k230-rst
Merge tag 'aspeed-6.17-drivers-1' of https://git.kernel.org/pub/scm/linux/kernel/git/bmc/linux into soc/drivers
ASPEED SoC driver updates for 6.17
The ASPEED LPC snoop driver was recently the cause of some concern. In addition
to the initial fixes, the channel configuration paths are refactored to improve
robustness against errors.
* tag 'aspeed-6.17-drivers-1' of https://git.kernel.org/pub/scm/linux/kernel/git/bmc/linux:
soc: aspeed: lpc-snoop: Lift channel config to const structs
soc: aspeed: lpc-snoop: Consolidate channel initialisation
soc: aspeed: lpc-snoop: Use dev_err_probe() where possible
soc: aspeed: lpc-snoop: Switch to devm_clk_get_enabled()
soc: aspeed: lpc-snoop: Rearrange channel paths
soc: aspeed: lpc-snoop: Rename 'channel' to 'index' in channel paths
soc: aspeed: lpc-snoop: Constrain parameters in channel paths
soc: aspeed: lpc-snoop: Ensure model_data is valid
soc: aspeed: lpc-snoop: Don't disable channels that aren't enabled
soc: aspeed: lpc-snoop: Cleanup resources in stack-order
Merge tag 'scmi-updates-6.17' of https://git.kernel.org/pub/scm/linux/kernel/git/sudeep.holla/linux into soc/drivers
Arm SCMI updates for v6.17
1. A fix is introduced to correct turbo frequency marking for 64-bit
devices with sustained frequencies over 4GHz, ensuring accurate turbo
frequency identification.
2. Debug capabilities are being improved by introducing in-flight transfer
tracking using debug counters, which help diagnose transfer congestion
and behavior. Additional tracepoints are added to log in-flight counts
at transfer begin and end, offering better runtime insight. The debug
counters now support decrement operations using a newly added
scmi_dec_count helper, making counter tracking symmetric and more robust.
3. A race condition in suspend-resume logic is being resolved by ensuring
SCMI_SYSPOWER_IDLE state is set early during resume, improving suspend
reliability under certain conditions. New suspend and resume operations
are added to the scmi_bus_type to enable finer power management control
for SCMI-based devices.
4. Finally enhancements are also made to avoid registering notifiers for
events that a platform does not support, reducing unnecessary overhead
by checking for unsupported event types during protocolinitialization.
* tag 'scmi-updates-6.17' of https://git.kernel.org/pub/scm/linux/kernel/git/sudeep.holla/linux:
firmware: arm_scmi: Convert to SYSTEM_SLEEP_PM_OPS
firmware: arm_scmi: Avoid notifier registration for unsupported events
firmware: arm_scmi: power_control: Ensure SCMI_SYSPOWER_IDLE is set early during resume
firmware: arm_scmi: Add power management operations to SCMI bus
include: trace: Add tracepoint support for inflight xfer count
firmware: arm_scmi: Track number of inflight SCMI transfers
firmware: arm_scmi: Add support for debug counter decrement
firmware: arm_scmi: Fix up turbo frequencies selection
Merge tag 'smccc-updates-6.17' of https://git.kernel.org/pub/scm/linux/kernel/git/sudeep.holla/linux into soc/drivers
firmware: smccc: Update for v6.17
Just a single update extending arm_smccc_hypervisor_has_uuid() to
support both SMC and HVC conduits, enabling UUID retrieval when Linux
runs as the root partition under Microsoft Hypervisor (MSHV), where
SMC is used. It leverages arm_smccc_1_1_invoke() to dynamically select
the appropriate conduit, enhancing compatibility beyond just HVC.
* tag 'smccc-updates-6.17' of https://git.kernel.org/pub/scm/linux/kernel/git/sudeep.holla/linux:
firmware: smccc: Support both smc and hvc conduits for getting hyp UUID
Add support to enable Memory Controller (MC) and External Memory
Controller (EMC) drivers for Tegra264. The nodes for MC and EMC are
mostly the same as Tegra234 but differ in number of channels and
interrupt numbers.
The patch also adds the bandwidth manager definitions required for
Tegra264 and uses them to populate the memory client table. All of
these are needed to properly enable memory interconnect (ICC) support.
Add bindings for the Memory Controller (MC) and External Memory
Controller (EMC) found on the Tegra264 SoC. Tegra264 SoC has a different
number of interrupt lines for MC sub-units: UCF_SOC, hub, hub common,
syncpoint and MC channel. The total number of interrupt lines is eight.
Update maxItems for MC interrupts accordingly.
This also adds a header containing the memory client ID definitions that
are used by the interconnects property in DT and the tegra_mc_client
table in the MC driver. These IDs are defined by the hardware, so the
numbering doesn't start at 0 and contains holes. Also added are the
stream IDs for various hardware blocks found on Tegra264. These are
allocated as blocks of 256 IDs and each block can be subdivided for
additional fine-grained isolation if needed.
firmware: tegra: bpmp: Fix build failure for tegra264-only config
The definition of tegra186_bpmp_ops was not updated in sync with the use
in bpmp.c:
drivers/firmware/tegra/bpmp.c:856:17: error: 'tegra186_bpmp_ops' undeclared here (not in a function); did you mean 'tegra_bpmp_ops'?
856 | .ops = &tegra186_bpmp_ops,
aarch64-linux-ld: drivers/firmware/tegra/bpmp.o:(.rodata+0x2f0): undefined reference to `tegra186_bpmp_ops'
Update the Makefile as needed.
There is really no need to hide the declaration based on the configuration,
so just expose it unconditionally so it never has to be updated again
for the next SoC.
bus: imx-aipstz: allow creating pdevs for child buses
devm_of_platform_populate() passes a NULL as the bus OF match table
to the underlying of_platform_populate(), meaning child bus devices
of the AIPSTZ bridge will not have its children devices created. Since
some SoCs (e.g. i.MX8MP) use this particular setup (e.g. SPBA bus, which
is a child of AIPSTZ5 and has multiple child nodes), the driver needs to
support it.
Therefore, replace devm_of_platform_populate() with of_platform_populate()
and pass a reference to the bus OF match table to it. For now, the only
possible child buses are simple buses.
Since the usage of devres is dropped, the complementary operation of
of_platform_populate() needs to be called during the driver's removal.
Signed-off-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com> Fixes: 796cba2dd4d9 ("bus: add driver for IMX AIPSTZ bridge") Reported-by: Alexander Stein <alexander.stein@ew.tq-group.com> Closes: https://lore.kernel.org/lkml/5029548.31r3eYUQgx@steina-w/#t Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org>
soc/tegra: cbb: Add support for CBB fabrics in Tegra254
Add support for CBB 2.0 based fabrics in Tegra254 SoC using ACPI.
Fabrics reporting errors are: C2C, GPU and Display_Cluster. Tegra254
uses a hardware based lookup to get target node address, so the
target_map tables for each fabric are no longer needed.
soc/tegra: cbb: Support HW lookup to get timed out target address
Add support for hardware based lookup to get the address of the timed
out target node. This features is added in upcoming SoCs and avoids the
need for creating per fabric target_map tables in the driver.
soc/tegra: cbb: Improve handling for per SoC fabric data
Improve handling for the per SoC fabrics and targets.
The below changes make them more flexible and ready for future SoC's.
- Added SoC prefix to Fabric_ID enums.
- Rename *lookup_target_timeout() to *sw_lookup_target_timeout() to
make it separate from HW based lookup function to be added later.
- Moved target_map within fabric_lookup table to make it easy to
check whether SW vs HW lookup is supported and handle accordingly.
- Slight improvements to some error prints.
soc/tegra: cbb: Make error interrupt enable and status per SoC
Make the error interrupt enable and error status fields as per SoC. Both
of these fields can change for different SoC's. Moving them to per SoC
data helps to set or clear the required bits only for a SoC.
soc/tegra: cbb: Clear ERR_FORCE register with ERR_STATUS
When error is injected with the ERR_FORCE register, then this register
is not auto cleared on clearing the ERR_STATUS register. This causes
repeated interrupts on error injection. To fix, set the ERR_FORCE to
zero along with clearing the ERR_STATUS register after handling error.
firmware: arm_scmi: Convert to SYSTEM_SLEEP_PM_OPS
The old SET_SYSTEM_SLEEP_PM_OPS() macro leads to a warning about an
unused function:
| drivers/firmware/arm_scmi/scmi_power_control.c:363:12: error:
| 'scmi_system_power_resume' defined but not used [-Werror=unused-function]
| static int scmi_system_power_resume(struct device *dev)
The proper way to do this these days is to use SYSTEM_SLEEP_PM_OPS()
and pm_sleep_ptr().
Fixes: 9a0658d3991e ("firmware: arm_scmi: power_control: Ensure SCMI_SYSPOWER_IDLE is set early during resume") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Peng Fan <peng.fan@nxp.com>
Message-Id: <20250709070107.1388512-1-arnd@kernel.org> Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
firmware: arm_scmi: Avoid notifier registration for unsupported events
Some platforms may be configured to not support notification events from
certain sources. This scenario is already handled gracefully by avoiding
any attempt to send a notification enable request for those sources, as
such requests would inevitably fail.
However, in a more extreme case, a platform might not support even a
single source for a given event type. In this situation, allowing
notifier registration is meaningless. Attempting to register a notifier
would serve no purpose and only result in unnecessary overhead.
To address this, we now detect such conditions during the protocol
initialization. When identified, we flag the unsupported event types and
reject any subsequent notifier registration attempts for them with
-ENOTSUPP. This early rejection avoids redundant processing and
simplifies runtime logic.
Renesas Kconfig is using "SoC serial number" for CONFIG symbol, but is
using "SoC chip name" for menu description. Because of it, it looks
random order when we run "make menuconfig".
commit 6d5aded8d57fc ("soc: renesas: Sort driver description title")
sorted Renesas Kconfig by menu description title order, but it makes
confusable to add new config.
Let's unify "ARMxx Platform support for ${CHIP_NUMBER} (${CHIP_NAME}),
and sort it again.
irq_domain_create_simple() takes fwnode as the first argument. It can be
extracted from the struct device using dev_fwnode() helper instead of
using of_node with of_fwnode_handle().
Andrew Jeffery [Mon, 16 Jun 2025 13:13:47 +0000 (22:43 +0930)]
soc: aspeed: lpc-snoop: Lift channel config to const structs
The shifts and masks for each channel are defined by hardware and
are not something that changes at runtime. Accordingly, describe the
information in an array of const structs and associate elements with
each channel instance, removing the need for the switch and handling of
its default case.
Previously, channel initialisation was a bit perilous with respect to
resource cleanup in error paths. While the implementation had issues,
it at least made an effort to eliminate some of its problems by first
testing whether any channels were enabled, and bailing out if not.
Having improved the robustness of resource handling in probe() we can
now rearrange the initial channel test to be located with the subsequent
test, and rework the unrolled conditional logic to use a loop for an
improvement in readability.
Andrew Jeffery [Mon, 16 Jun 2025 13:13:43 +0000 (22:43 +0930)]
soc: aspeed: lpc-snoop: Rearrange channel paths
Order assignments such that tests for conditions not involving resource
acquisition are ordered before those testing acquired resources, and
order managed resource acquisition before unmanaged where possible. This
way we minimise the amount of manual cleanup required.
In the process, improve readability of the code by introducing a channel
pointer that takes the place of the repeated object lookups.
Andrew Jeffery [Mon, 16 Jun 2025 13:13:40 +0000 (22:43 +0930)]
soc: aspeed: lpc-snoop: Ensure model_data is valid
of_device_get_match_data() can return NULL, though shouldn't in current
circumstances. Regardless, initialise model_data closer to use so it's
clear we need to test for validity prior to dereferencing.
Peng Fan [Fri, 4 Jul 2025 03:09:36 +0000 (11:09 +0800)]
firmware: arm_scmi: power_control: Ensure SCMI_SYSPOWER_IDLE is set early during resume
Fix a race condition where a second suspend notification from another
SCMI agent wakes the system before SCMI_SYSPOWER_IDLE is set, leading
to ignored suspend requests. This is due to interrupts triggering early
execution of `scmi_userspace_notifier()` before the SCMI state is updated.
To resolve this, set SCMI_SYSPOWER_IDLE earlier in the device resume
path, prior to `thaw_processes()`. This ensures the SCMI state is
correct when the notifier runs, allowing the system to suspend again
as expected.
On some platforms using SCMI, SCP cannot distinguish between CPU idle
and suspend since both result in cluster power-off. By explicitly setting
the idle state early, the Linux SCMI agent can correctly re-suspend in
response to external notifications.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Message-Id: <20250704-scmi-pm-v2-2-9316cec2f9cc@nxp.com> Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Peng Fan [Fri, 4 Jul 2025 03:09:35 +0000 (11:09 +0800)]
firmware: arm_scmi: Add power management operations to SCMI bus
Introduce suspend and resume power management callbacks for `scmi_bus_type`,
modeled after `platform_pm_ops`. This enables SCMI devices on the bus to
implement their own suspend and resume behavior, allowing for more
fine-grained power control at the device level.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Message-Id: <20250704-scmi-pm-v2-1-9316cec2f9cc@nxp.com> Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Thierry Reding [Tue, 6 May 2025 13:31:16 +0000 (15:31 +0200)]
firmware: tegra: Fix IVC dependency problems
The IVC code is library code that other drivers need to select if they
need that library. However, if the symbol is user-selectable this can
lead to conflicts.
Fix this by making the symbol only selectable for COMPILE_TEST and add
a select TEGRA_IVC to TEGRA_BPMP, which is currently the only user.
Thierry Reding [Tue, 6 May 2025 13:31:14 +0000 (15:31 +0200)]
soc/tegra: pmc: Add Tegra264 support
The PMC block on Tegra264 has undergone a few small changes since it's
Tegra234 predecessor. Match on the new compatible string to select the
updated SoC-specific data.
Merge tag 'renesas-drivers-for-v6.17-tag1' of https://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-devel into soc/drivers
Renesas driver updates for v6.17
- Initial support for the Renesas RZ/T2H (R9A09G077) and RZ/N2H
(R9A09G087) SoCs,
- Convert the RZ/V2M External Power Sequence Controller (PWC) driver
to the new GPIO line value setter callbacks.
* tag 'renesas-drivers-for-v6.17-tag1' of https://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-devel:
soc: renesas: pwc-rzv2m: Use new GPIO line value setter callbacks
soc: renesas: Add RZ/N2H (R9A09G087) config option
soc: renesas: Add RZ/T2H (R9A09G077) config option
Philip Radford [Mon, 30 Jun 2025 10:55:42 +0000 (10:55 +0000)]
firmware: arm_scmi: Track number of inflight SCMI transfers
Add a new debug counter, `XFERS_INFLIGHT`, to track the number of
currently active in-flight SCMI message transfers. This helps in
understanding system behavior and diagnosing potential issues with
pending or stuck messages.
The counter is incremented when a transfer is registered as in-flight,
and decremented when it completes and is released.
It is automatically added to debugfs for visibility through SCMI debugfs.
Philip Radford [Mon, 30 Jun 2025 10:55:41 +0000 (10:55 +0000)]
firmware: arm_scmi: Add support for debug counter decrement
Introduce a new `scmi_dec_count` helper to complement the existing
`scmi_inc_count`, allowing controlled decrement of SCMI debug counters.
This provides symmetry in debug counter management and enables
accurate tracking of counters that may both increase and decrease,
such as those used for in-flight message tracking.
Only active when CONFIG_ARM_SCMI_DEBUG_COUNTERS is enabled.
Andrew Jeffery [Mon, 16 Jun 2025 13:13:38 +0000 (22:43 +0930)]
soc: aspeed: lpc-snoop: Cleanup resources in stack-order
Free the kfifo after unregistering the miscdev in
aspeed_lpc_disable_snoop() as the kfifo is initialised before the
miscdev in aspeed_lpc_enable_snoop().
The secure AHB to IP Slave (AIPSTZ) bus bridge provides access control
configurations meant to restrict access to certain peripherals.
Some of the configurations include:
1) Marking masters as trusted for R/W. Based on this
(and the configuration of the accessed peripheral), the bridge
may choose to abort the R/W transactions issued by certain
masters.
2) Allowing/disallowing write accesses to peripherals.
Add driver for this IP. Since there's currently no framework for
access controllers (and since there's currently no need for having
flexibility w.r.t the configurations) all this driver does is it
applies a relaxed, "default" configuration, in which all masters
are trusted for R/W.
Note that some instances of this IP (e.g: AIPSTZ5 on i.MX8MP) may be tied
to a power domain and may lose their configuration when the domain is
powered off. This is why the configuration has to be restored when the
domain is powered on.
Co-developed-by: Daniel Baluta <daniel.baluta@nxp.com> Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com> Signed-off-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Peter Robinson [Mon, 30 Jun 2025 17:52:58 +0000 (18:52 +0100)]
reset: brcmstb: Enable reset drivers for ARCH_BCM2835
The BRCMSTB and BRCMSTB_RESCAL reset drivers are also
used in the BCM2712, AKA the RPi5. The RPi platforms
have typically used the ARCH_BCM2835, and the PCIe
support for this SoC can use this config which depends
on these drivers so enable building them when just that
arch option is enabled to ensure the platform works as
expected.
Lad Prabhakar [Wed, 28 May 2025 13:30:31 +0000 (14:30 +0100)]
dt-bindings: reset: renesas,rzv2h-usb2phy: Document RZ/V2N SoC support
Document support for the USB2PHY reset controller found on the Renesas
RZ/V2N (R9A09G056) SoC. The reset controller IP is functionally identical
to that on the RZ/V2H(P) SoC, so no driver changes are needed. The existing
`renesas,r9a09g057-usb2phy-reset` compatible will be used as a fallback
for the RZ/V2N SoC.
pinctrl: rp1: Implement RaspberryPi RP1 pinmux/pinconf support
The current implementation for the pin controller peripheral
on the RP1 chipset supports gpio functionality and just the
basic configuration of pin hw capabilities.
Add support for selecting the pin alternate function (pinmux)
and full configuration of the pin (pinconf).
Jason-JH Lin [Tue, 24 Jun 2025 10:39:03 +0000 (18:39 +0800)]
soc: mediatek: mtk-mutex: Fix confusing usage of MUTEX_MOD2
The usage of MUTEX_MOD1 and MUTEX_MOD2 for calculating mod settings
over 32 has been confusing. To improve consistency and clarity, these
defines need to fit into the same MUTEX_MOD define as possible.
However, MUTEX_MOD1 cannot be directly used for all SoCs because,
for example, the mod1 register (0x34) of MT2712 is not adjacent to
its mod0 register (0x2c). To address this, a `mutex_mod1_reg` field
is introduced in the mutex driver data structure. This allows all
SoCs to use a unified MUTEX_MOD to determine their register offsets.
With this change, the separate usage of MUTEX_MOD1 and MUTEX_MOD2 is
eliminated, simplifying the logic for obtaining offsets and mod IDs.
Signed-off-by: Jason-JH Lin <jason-jh.lin@mediatek.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20250624103928.408194-1-jason-jh.lin@mediatek.com Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Sibi Sankar [Wed, 14 May 2025 21:47:19 +0000 (03:17 +0530)]
firmware: arm_scmi: Fix up turbo frequencies selection
Sustained frequency when greater than or equal to 4Ghz on 64-bit devices
currently result in marking all frequencies as turbo. Address the turbo
frequency selection bug by fixing the truncation.
Fixes: a897575e79d7 ("firmware: arm_scmi: Add support for marking certain frequencies as turbo") Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com>
Message-Id: <20250514214719.203607-1-quic_sibis@quicinc.com> Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Linus Torvalds [Sun, 22 Jun 2025 17:50:36 +0000 (10:50 -0700)]
Merge tag 'i2c-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
- subsystem: convert drivers to use recent callbacks of struct
i2c_algorithm A typical after-rc1 cleanup, which I couldn't send in
time for rc2
- tegra: fix YAML conversion of device tree bindings
- k1: re-add a check which got lost during upstreaming
* tag 'i2c-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: k1: check for transfer error
i2c: use inclusive callbacks in struct i2c_algorithm
dt-bindings: i2c: nvidia,tegra20-i2c: Specify the required properties
Linus Torvalds [Sun, 22 Jun 2025 17:30:44 +0000 (10:30 -0700)]
Merge tag 'x86_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Borislav Petkov:
- Make sure the array tracking which kernel text positions need to be
alternatives-patched doesn't get mishandled by out-of-order
modifications, leading to it overflowing and causing page faults when
patching
- Avoid an infinite loop when early code does a ranged TLB invalidation
before the broadcast TLB invalidation count of how many pages it can
flush, has been read from CPUID
- Fix a CONFIG_MODULES typo
- Disable broadcast TLB invalidation when PTI is enabled to avoid an
overflow of the bitmap tracking dynamic ASIDs which need to be
flushed when the kernel switches between the user and kernel address
space
- Handle the case of a CPU going offline and thus reporting zeroes when
reading top-level events in the resctrl code
* tag 'x86_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/alternatives: Fix int3 handling failure from broken text_poke array
x86/mm: Fix early boot use of INVPLGB
x86/its: Fix an ifdef typo in its_alloc()
x86/mm: Disable INVLPGB when PTI is enabled
x86,fs/resctrl: Remove inappropriate references to cacheinfo in the resctrl subsystem
Linus Torvalds [Sun, 22 Jun 2025 17:17:51 +0000 (10:17 -0700)]
Merge tag 'irq_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq fixes from Borislav Petkov:
- Fix missing prototypes warnings
- Properly initialize work context when allocating it
- Remove a method tracking when managed interrupts are suspended during
hotplug, in favor of the code using a IRQ disable depth tracking now,
and have interrupts get properly enabled again on restore
- Make sure multiple CPUs getting hotplugged don't cause wrong tracking
of the managed IRQ disable depth
* tag 'irq_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip/ath79-misc: Fix missing prototypes warnings
genirq/irq_sim: Initialize work context pointers properly
genirq/cpuhotplug: Restore affinity even for suspended IRQ
genirq/cpuhotplug: Rebalance managed interrupts across multi-CPU hotplug
Linus Torvalds [Sun, 22 Jun 2025 17:11:45 +0000 (10:11 -0700)]
Merge tag 'perf_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Borislav Petkov:
- Avoid a crash on a heterogeneous machine where not all cores support
the same hw events features
- Avoid a deadlock when throttling events
- Document the perf event states more
- Make sure a number of perf paths switching off or rescheduling events
call perf_cgroup_event_disable()
- Make sure perf does task sampling before its userspace mapping is
torn down, and not after
* tag 'perf_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel: Fix crash in icl_update_topdown_event()
perf: Fix the throttle error of some clock events
perf: Add comment to enum perf_event_state
perf/core: Fix WARN in perf_cgroup_switch()
perf: Fix dangling cgroup pointer in cpuctx
perf: Fix cgroup state vs ERROR
perf: Fix sample vs do_exit()
Linus Torvalds [Sun, 22 Jun 2025 17:09:23 +0000 (10:09 -0700)]
Merge tag 'locking_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking fixes from Borislav Petkov:
- Make sure the switch to the global hash is requested always under a
lock so that two threads requesting that simultaneously cannot get to
inconsistent state
- Reject negative NUMA nodes earlier in the futex NUMA interface
handling code
- Selftests fixes
* tag 'locking_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Verify under the lock if hash can be replaced
futex: Handle invalid node numbers supplied by user
selftests/futex: Set the home_node in futex_numa_mpol
selftests/futex: getopt() requires int as return value.
* tag 'edac_urgent_for_v6.16_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
EDAC/igen6: Fix NULL pointer dereference
EDAC/amd64: Correct number of UMCs for family 19h models 70h-7fh
Linus Torvalds [Sun, 22 Jun 2025 16:46:11 +0000 (09:46 -0700)]
Merge tag 'v6.16-rc2-smb3-client-fixes-v2' of git://git.samba.org/sfrench/cifs-2.6
Pull smb client fixes from Steve French:
- Multichannel channel allocation fix for Kerberos mounts
- Two reconnect fixes
- Fix netfs_writepages crash with smbdirect/RDMA
- Directory caching fix
- Three minor cleanup fixes
- Log error when close cached dirs fails
* tag 'v6.16-rc2-smb3-client-fixes-v2' of git://git.samba.org/sfrench/cifs-2.6:
smb: minor fix to use SMB2_NTLMV2_SESSKEY_SIZE for auth_key size
smb: minor fix to use sizeof to initialize flags_string buffer
smb: Use loff_t for directory position in cached_dirents
smb: Log an error when close_all_cached_dirs fails
cifs: Fix prepare_write to negotiate wsize if needed
smb: client: fix max_sge overflow in smb_extract_folioq_to_rdma()
smb: client: fix first command failure during re-negotiation
cifs: Remove duplicate fattr->cf_dtype assignment from wsl_to_fattr() function
smb: fix secondary channel creation issue with kerberos by populating hostname when adding channels
Alex Elder [Mon, 16 Jun 2025 12:51:36 +0000 (07:51 -0500)]
i2c: k1: check for transfer error
If spacemit_i2c_xfer_msg() times out waiting for a message transfer to
complete, or if the hardware reports an error, it returns a negative
error code (-ETIMEDOUT, -EAGAIN, -ENXIO. or -EIO).
The sole caller of spacemit_i2c_xfer_msg() is spacemit_i2c_xfer(),
which is the i2c_algorithm->xfer callback function. It currently
does not save the value returned by spacemit_i2c_xfer_msg().
The result is that transfer errors go unreported, and a caller
has no indication anything is wrong.
When this code was out for review, the return value *was* checked
in early versions. But for some reason, that assignment got dropped
between versions 5 and 6 of the series, perhaps related to reworking
the code to merge spacemit_i2c_xfer_core() into spacemit_i2c_xfer().
Simply assigning the value returned to "ret" fixes the problem.
Fixes: 5ea558473fa31 ("i2c: spacemit: add support for SpacemiT K1 SoC") Signed-off-by: Alex Elder <elder@riscstar.com> Cc: <stable@vger.kernel.org> # v6.15+ Reviewed-by: Troy Mitchell <troymitchell988@gmail.com> Link: https://lore.kernel.org/r/20250616125137.1555453-1-elder@riscstar.com Signed-off-by: Andi Shyti <andi@smida.it> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Linus Torvalds [Sat, 21 Jun 2025 16:20:15 +0000 (09:20 -0700)]
Merge tag 'nfsd-6.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux
Pull nfsd fixes from Chuck Lever:
- Two fixes for commits in the nfsd-6.16 merge
- One fix for the recently-added NFSD netlink facility
- One fix for a remote SunRPC crasher
* tag 'nfsd-6.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux:
sunrpc: handle SVC_GARBAGE during svc auth processing as auth error
nfsd: use threads array as-is in netlink interface
SUNRPC: Cleanup/fix initial rq_pages allocation
NFSD: Avoid corruption of a referring call list
Bharath SM [Thu, 19 Jun 2025 15:35:34 +0000 (21:05 +0530)]
smb: minor fix to use SMB2_NTLMV2_SESSKEY_SIZE for auth_key size
Replaced hardcoded value 16 with SMB2_NTLMV2_SESSKEY_SIZE
in the auth_key definition and memcpy call.
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Signed-off-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Bharath SM [Thu, 19 Jun 2025 15:35:33 +0000 (21:05 +0530)]
smb: minor fix to use sizeof to initialize flags_string buffer
Replaced hardcoded length with sizeof(flags_string).
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Signed-off-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Bharath SM [Thu, 19 Jun 2025 15:35:32 +0000 (21:05 +0530)]
smb: Use loff_t for directory position in cached_dirents
Change the pos field in struct cached_dirents from int to loff_t
to support large directory offsets. This avoids overflow and
matches kernel conventions for directory positions.
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Signed-off-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Paul Aurich [Wed, 20 Nov 2024 16:01:54 +0000 (08:01 -0800)]
smb: Log an error when close_all_cached_dirs fails
Under low-memory conditions, close_all_cached_dirs() can't move the
dentries to a separate list to dput() them once the locks are dropped.
This will result in a "Dentry still in use" error, so add an error
message that makes it clear this is what happened:
[ 495.281119] CIFS: VFS: \\otters.example.com\share Out of memory while dropping dentries
[ 495.281595] ------------[ cut here ]------------
[ 495.281887] BUG: Dentry ffff888115531138{i=78,n=/} still in use (2) [unmount of cifs cifs]
[ 495.282391] WARNING: CPU: 1 PID: 2329 at fs/dcache.c:1536 umount_check+0xc8/0xf0
Also, bail out of looping through all tcons as soon as a single
allocation fails, since we're already in trouble, and kmalloc() attempts
for subseqeuent tcons are likely to fail just like the first one did.
Signed-off-by: Paul Aurich <paul@darkrain42.org> Acked-by: Bharath SM <bharathsm@microsoft.com> Suggested-by: Ruben Devos <rdevos@oxya.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
David Howells [Wed, 18 Jun 2025 15:39:47 +0000 (16:39 +0100)]
cifs: Fix prepare_write to negotiate wsize if needed
Fix cifs_prepare_write() to negotiate the wsize if it is unset.
Reviewed-by: Shyam Prasad N <nspmangalore@gmail.com> Reviewed-by: Bharath SM <bharathsm@microsoft.com> Signed-off-by: David Howells <dhowells@redhat.com>
cc: Paulo Alcantara <pc@manguebit.org>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-cifs@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
cc: Tom Talpey <tom@talpey.com>
cc: linux-cifs@vger.kernel.org Reviewed-by: David Howells <dhowells@redhat.com> Reviewed-by: Tom Talpey <tom@talpey.com> Fixes: c45ebd636c32 ("cifs: Provide the capability to extract from ITER_FOLIOQ to RDMA SGEs") Signed-off-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Steve French <stfrench@microsoft.com>
zhangjian [Thu, 19 Jun 2025 01:18:29 +0000 (09:18 +0800)]
smb: client: fix first command failure during re-negotiation
after fabc4ed200f9, server_unresponsive add a condition to check whether client
need to reconnect depending on server->lstrp. When client failed to reconnect
for some time and abort connection, server->lstrp is updated for the last time.
In the following scene, server->lstrp is too old. This cause next command
failure in re-negotiation rather than waiting for re-negotiation done.
1. mount -t cifs -o username=Everyone,echo_internal=10 //$server_ip/export /mnt
2. ssh $server_ip "echo b > /proc/sysrq-trigger &"
3. ls /mnt
4. sleep 21s
5. ssh $server_ip "service firewalld stop"
6. ls # return EHOSTDOWN
If the interval between 5 and 6 is too small, 6 may trigger sending negotiation
request. Before backgrounding cifsd thread try to receive negotiation response
from server in cifs_readv_from_socket, server_unresponsive may trigger
cifs_reconnect which cause 6 to be failed:
ls thread
----------------
smb2_negotiate
server->tcpStatus = CifsInNegotiate
compound_send_recv
wait_for_compound_request
ls thread
----------------
cifs_sync_mid_result return EAGAIN
smb2_negotiate return EHOSTDOWN
Though server->lstrp means last server response time, it is updated in
cifs_abort_connection and cifs_get_tcp_session. We can also update server->lstrp
before switching into CifsInNegotiate state to avoid failure in 6.
Fixes: 7ccc1465465d ("smb: client: fix hang in wait_for_response() for negproto") Acked-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Acked-by: Meetakshi Setiya <msetiya@microsoft.com> Signed-off-by: zhangjian <zhangjian496@huawei.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Linus Torvalds [Sat, 21 Jun 2025 15:27:12 +0000 (08:27 -0700)]
Merge tag 'acpi-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI fix from Rafael Wysocki:
"Fix a crash in ACPICA while attempting to evaluate a control method
that expects more arguments than are being passed to it, which was
exposed by a defective firmware update from a prominent OEM on
multiple systems (Rafael Wysocki)"
* tag 'acpi-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPICA: Refuse to evaluate a method if arguments are missing
Linus Torvalds [Sat, 21 Jun 2025 15:21:10 +0000 (08:21 -0700)]
Merge tag 'pci-v6.16-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci
Pull PCI fixes from Bjorn Helgaas:
- Set up runtime PM even for devices that lack a PM Capability as we
did before 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when
initializing"), which broke resume in some VFIO scenarios (Mario
Limonciello)
- Ignore pciehp Presence Detect Changed events caused by DPC, even if
they occur after a Data Link Layer State Changed event, to fix a VFIO
GPU passthrough regression in v6.13 (Lukas Wunner)
* tag 'pci-v6.16-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
PCI: pciehp: Ignore belated Presence Detect Changed caused by DPC
PCI/PM: Set up runtime PM even for devices without PCI PM
Linus Torvalds [Sat, 21 Jun 2025 14:59:45 +0000 (07:59 -0700)]
Merge tag 'perf-tools-fixes-for-v6.16-1-2025-06-20' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools
Pull perf tools fixes from Arnaldo Carvalho de Melo:
- Fix some file descriptor leaks that stand out with recent changes to
'perf list'
- Fix prctl include to fix building 'perf bench futex' hash with musl
libc
- Restrict 'perf test' uniquifying entry to machines with 'uncore_imc'
PMUs
- Document new output fields (op, cache, mem, dtlb, snoop) used with
'perf mem'
- Synchronize kernel header copies
* tag 'perf-tools-fixes-for-v6.16-1-2025-06-20' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools:
tools headers x86 cpufeatures: Sync with the kernel sources
perf bench futex: Fix prctl include in musl libc
perf test: Directory file descriptor leak
perf evsel: Missed close() when probing hybrid core PMUs
tools headers: Synchronize linux/bits.h with the kernel sources
tools arch amd ibs: Sync ibs.h with the kernel sources
tools arch x86: Sync the msr-index.h copy with the kernel sources
tools headers: Syncronize linux/build_bug.h with the kernel sources
tools headers: Update the copy of x86's mem{cpy,set}_64.S used in 'perf bench'
tools headers UAPI: Sync linux/kvm.h with the kernel sources
tools headers UAPI: Sync the drm/drm.h with the kernel sources
perf beauty: Update copy of linux/socket.h with the kernel sources
tools headers UAPI: Sync kvm header with the kernel sources
tools headers x86 svm: Sync svm headers with the kernel sources
tools headers UAPI: Sync KVM's vmx.h header with the kernel sources
tools kvm headers arm64: Update KVM header from the kernel sources
tools headers UAPI: Sync linux/prctl.h with the kernel sources to pick FUTEX knob
perf mem: Document new output fields (op, cache, mem, dtlb, snoop)
tools headers: Update the fs headers with the kernel sources
perf test: Restrict uniquifying test to machines with 'uncore_imc'
Linus Torvalds [Sat, 21 Jun 2025 05:36:48 +0000 (22:36 -0700)]
Merge tag 'mtd/fixes-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux
Pull mtd fixes from Miquel Raynal:
"The main fix that really needs to get in is the revert of the patch
adding the new mtd_master class, because it entirely fails the
partitioning if a specific Kconfig option is set. We need to think how
to handle that differently, so let's revert it as we need to get back
to the pen and paper situation again.
Otherwise the definition of some Winbond SPI NAND chips are receiving
some fixes (geometry and maximum frequency, mostly).
And finally a small memory leak gets also fixed"
* tag 'mtd/fixes-for-6.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux:
mtd: spinand: fix memory leak of ECC engine conf
mtd: spinand: winbond: Prevent unsupported frequencies on dual/quad I/O variants
mtd: spinand: winbond: Increase maximum frequency on an octal operation
mtd: spinand: winbond: Fix W35N number of planes/LUN
Revert "mtd: core: always create master device"
Currently the call_rcu() API does not check whether a callback
pointer is NULL. If NULL is passed, rcu_core() will try to invoke
it, resulting in NULL pointer dereference and a kernel crash.
To prevent this and improve debuggability, this patch adds a check
for NULL and emits a kernel stack trace to help identify a faulty
caller.
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Reviewed-by: Joel Fernandes <joelagnelf@nvidia.com> Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
Binbin Wu [Tue, 10 Jun 2025 02:14:21 +0000 (10:14 +0800)]
KVM: TDX: Exit to userspace for GetTdVmCallInfo
Exit to userspace for TDG.VP.VMCALL<GetTdVmCallInfo> via KVM_EXIT_TDX,
to allow userspace to provide information about the support of
TDVMCALLs when r12 is 1 for the TDVMCALLs beyond the GHCI base API.
GHCI spec defines the GHCI base TDVMCALLs: <GetTdVmCallInfo>, <MapGPA>,
<ReportFatalError>, <Instruction.CPUID>, <#VE.RequestMMIO>,
<Instruction.HLT>, <Instruction.IO>, <Instruction.RDMSR> and
<Instruction.WRMSR>. They must be supported by VMM to support TDX guests.
For GetTdVmCallInfo
- When leaf (r12) to enumerate TDVMCALL functionality is set to 0,
successful execution indicates all GHCI base TDVMCALLs listed above are
supported.
Update the KVM TDX document with the set of the GHCI base APIs.
- When leaf (r12) to enumerate TDVMCALL functionality is set to 1, it
indicates the TDX guest is querying the supported TDVMCALLs beyond
the GHCI base TDVMCALLs.
Exit to userspace to let userspace set the TDVMCALL sub-function bit(s)
accordingly to the leaf outputs. KVM could set the TDVMCALL bit(s)
supported by itself when the TDVMCALLs don't need support from userspace
after returning from userspace and before entering guest. Currently, no
such TDVMCALLs implemented, KVM just sets the values returned from
userspace.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
[Adjust userspace API. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
projects / thirdparty / linux.git / log
