Stephen Boyd [Mon, 6 Oct 2025 18:00:12 +0000 (13:00 -0500)]
Merge branches 'clk-microchip', 'clk-lookup' and 'clk-st' into clk-next
- Speed up clk_core_lookup() by using a hashtable
* clk-microchip:
ARM: at91: remove default values for PMC_PLL_ACR
clk: at91: add ACR in all PLL settings
clk: at91: sam9x7: Add peripheral clock id for pmecc
clk: at91: clk-master: Add check for divide by 3
clk: at91: clk-sam9x60-pll: force write to PLL_UPDT register
ARM: at91: pm: save and restore ACR during PLL disable/enable
* clk-lookup:
clk: Use hashtable for global clk lookups
clk: Sort include statements
* clk-st:
dt-bindings: stm32: cosmetic fixes for STM32MP25 clock and reset bindings
clk: stm32: introduce clocks for STM32MP21 platform
dt-bindings: stm32: add STM32MP21 clocks and reset bindings
Stephen Boyd [Mon, 6 Oct 2025 17:56:46 +0000 (12:56 -0500)]
Merge branches 'clk-samsung', 'clk-tegra' and 'clk-amlogic' into clk-next
* clk-samsung:
clk: s2mps11: add support for S2MPG10 PMIC clock
dt-bindings: clock: samsung,s2mps11: add s2mpg10
clk: samsung: exynos990: Add PERIC0 and PERIC1 clock support
dt-bindings: clock: exynos990: Add PERIC0 and PERIC1 clock units
clk: samsung: exynos990: Add missing USB clock registers to HSI0
clk: samsung: exynos990: Add LHS_ACEL gate clock for HSI0 and update CLK_NR_TOP
dt-bindings: clock: exynos990: Add LHS_ACEL clock ID for HSI0 block
clk: samsung: artpec-8: Add initial clock support for ARTPEC-8 SoC
clk: samsung: Add clock PLL support for ARTPEC-8 SoC
dt-bindings: clock: Add ARTPEC-8 clock controller
clk: samsung: exynos990: Add DPU_BUS and CMUREF mux/div and update CLKS_NR_TOP
dt-bindings: clock: exynos990: Extend clocks IDs
clk: samsung: exynos990: Replace bogus divs with fixed-factor clocks
clk: samsung: exynos990: Fix CMU_TOP mux/div bit widths
clk: samsung: exynos990: Use PLL_CON0 for PLL parent muxes
clk: samsung: pll: convert from round_rate() to determine_rate()
clk: samsung: cpu: convert from round_rate() to determine_rate()
clk: samsung: fsd: Add clk id for PCLK and PLL in CAM_CSI block
dt-bindings: clock: Add CAM_CSI clock macro for FSD
* clk-tegra:
clk: tegra: dfll: Add CVB tables for Tegra114
clk: tegra: Add DFLL DVCO reset control for Tegra114
dt-bindings: arm: tegra: Add ASUS TF101G and SL101
dt-bindings: reset: Add Tegra114 CAR header
dt-bindings: arm: tegra: Add Xiaomi Mi Pad (A0101)
dt-bindings: clock: tegra30: Add IDs for CSI pad clocks
dt-bindings: display: tegra: Move avdd-dsi-csi-supply from VI to CSI
dt-bindings: i2c: nvidia,tegra20-i2c: Document Tegra264 I2C
* clk-amlogic:
clk: amlogic: fix recent code refactoring
clk: amlogic: c3-peripherals: use helper for basic composite clocks
clk: amlogic: align s4 and c3 pwm clock descriptions
clk: amlogic: add composite clock helpers
clk: amlogic: use the common pclk definition
clk: amlogic: introduce a common pclk definition
clk: amlogic: pclk explicitly use CLK_IGNORE_UNUSED
clk: amlogic: drop CLK_SET_RATE_PARENT from peripheral clocks
clk: amlogic: move PCLK definition to clkc-utils
clk: amlogic: aoclk: use clkc-utils syscon probe
clk: amlogic: use probe helper in mmio based controllers
clk: amlogic: add probe helper for mmio based controllers
clk: amlogic: drop meson-clkcee
clk: amlogic: naming consistency alignment
clk: tegra: do not overallocate memory for bpmp clocks
struct tegra_bpmp::clocks is a pointer to a dynamically allocated array
of pointers to 'struct tegra_bpmp_clk'.
But the size of the allocated area is calculated like it is an array
containing actual 'struct tegra_bpmp_clk' objects - it's not true, there
are just pointers.
Found by Linux Verification Center (linuxtesting.org) with Svace static
analysis tool.
Fixes: 2db12b15c6f3 ("clk: tegra: Register clocks from root to leaf") Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Qianfeng Rong [Sat, 30 Aug 2025 12:27:52 +0000 (20:27 +0800)]
clk: ep93xx: Use int type to store negative error codes
Change the 'ret' variable in ep93xx_uart_clock_init() from unsigned int to
int, as it needs to store either negative error codes or zero.
Storing the negative error codes in unsigned type, doesn't cause an issue
at runtime but can be confusing. Additionally, assigning negative error
codes to unsigned type may trigger a GCC warning when the -Wsign-conversion
flag is enabled.
No effect on runtime.
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
André Draszik [Wed, 30 Jul 2025 09:31:35 +0000 (10:31 +0100)]
clk: s2mps11: add support for S2MPG10 PMIC clock
Add support for Samsung's S2MPG10 PMIC clock, which is similar to the
existing PMIC clocks supported by this driver.
S2MPG10 has three clock outputs @ 32kHz: AP, peri1 and peri2.
Acked-by: Stephen Boyd <sboyd@kernel.org> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by: André Draszik <andre.draszik@linaro.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
André Draszik [Wed, 30 Jul 2025 09:31:34 +0000 (10:31 +0100)]
dt-bindings: clock: samsung,s2mps11: add s2mpg10
The Samsung S2MPG10 clock controller is similar to the existing clock
controllers supported by this binding. Register offsets / layout are
slightly different, so it needs its own compatible.
Acked-by: Stephen Boyd <sboyd@kernel.org> Acked-by: Rob Herring (Arm) <robh@kernel.org> Signed-off-by: André Draszik <andre.draszik@linaro.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Gabriel Fernandez [Wed, 25 Jun 2025 09:07:25 +0000 (11:07 +0200)]
clk: stm32: introduce clocks for STM32MP21 platform
This driver is intended for the STM32MP21 clock family.
Signed-off-by: Nicolas Le Bayon <nicolas.le.bayon@foss.st.com> Reviewed-by: Alok Tiwari <alok.a.tiwari@oracle.com> Signed-off-by: Gabriel Fernandez <gabriel.fernandez@foss.st.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Gabriel Fernandez [Wed, 25 Jun 2025 09:07:24 +0000 (11:07 +0200)]
dt-bindings: stm32: add STM32MP21 clocks and reset bindings
Adds clock and reset binding entries for STM32MP21 SoC family.
Signed-off-by: Nicolas Le Bayon <nicolas.le.bayon@foss.st.com> Reviewed-by: Conor Dooley <conor.dooley@microchip.com> Signed-off-by: Gabriel Fernandez <gabriel.fernandez@foss.st.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Chen-Yu Tsai [Thu, 14 Aug 2025 03:53:16 +0000 (11:53 +0800)]
clk: Use hashtable for global clk lookups
A clk lookup using clk_core_lookup() is currently somewhat expensive
since it has to walk the whole clk tree to find a match. This is
extremely bad in the clk_core_init() function where it is used to look
for clk name conflicts, which is always the worst case of walking the
whole tree. Moreover, the number of clks checked increases as more
clks are registered, causing each subsequent clk registration becoming
slower.
Add a hashtable for doing clk lookups to replace the tree walk method.
On arm64 this increases kernel memory usage by 4 KB for the hashtable,
and 16 bytes (2 pointers) for |struct hlist_node| in each clk. On a
platform with around 800 clks, this reduces the time spent in
clk_core_lookup() significantly:
| PID 0 | kworker |
| before | after | before | after |
-------------------------------------------
avg | 203 us | 2.7 us | 123 us | 1.5 us |
-------------------------------------------
min | 4.7 us | 2.3 us | 102 us | 0.9 us |
-------------------------------------------
max | 867 us | 4.8 us | 237 us | 3.5 us |
-------------------------------------------
culm | 109 ms | 1.5 ms | 21 ms | 0.3 ms |
This in turn reduces the time spent in clk_hw_register(), and
ultimately, boot time. On a different system with close to 700 clks,
This reduces boot time by around 110 ms. While this doesn't seem like
a lot, this helps in cases where minimizing boot time is important.
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org> Tested-by: Brian Masney <bmasney@redhat.com> Reviewed-by: Brian Masney <bmasney@redhat.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Stephen Boyd [Sun, 21 Sep 2025 16:56:03 +0000 (09:56 -0700)]
Merge tag 'clk-microchip-6.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux into clk-microchip
Pull Microchip clk driver updates from Nicolas Ferre:
- add one clock for sam9x75
- new meaning for MCR register field in clk-master
- use force-write to PLL update register to ensure
reliable programming sequence
- update Analog Control Register (ACR) management to accommodate
differences across SoCs.
- ACR management dependency with one ARM PM patch added beforehand
* tag 'clk-microchip-6.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux:
ARM: at91: remove default values for PMC_PLL_ACR
clk: at91: add ACR in all PLL settings
clk: at91: sam9x7: Add peripheral clock id for pmecc
clk: at91: clk-master: Add check for divide by 3
clk: at91: clk-sam9x60-pll: force write to PLL_UPDT register
ARM: at91: pm: save and restore ACR during PLL disable/enable
st/stih407-clock.dtsi file has been removed in commit 65322c1daf51
("clk: st: flexgen: remove unused compatible"). This file has three
compatibles which are now dangling. Remove them from documentation.
Signed-off-by: Raphael Gallais-Pou <rgallaispou@gmail.com> Reviewed-by: Brian Masney <bmasney@redhat.com> Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com> Acked-by: Rob Herring (Arm) <robh@kernel.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Although minimizing the clock rate is the best for most scenarios, as
stated in commit 4d85abb0fb8e ("clk: bcm: rpi: Enable minimize for all
firmware clocks"), when it comes to the GPU, it's ideal to have the
maximum rate allowed.
Add an option to maximize a firmware clock's rate when the clock is
enabled and set this option for V3D.
Signed-off-by: Maíra Canal <mcanal@igalia.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
clk: bcm: rpi: Turn firmware clock on/off when preparing/unpreparing
Currently, when we prepare or unprepare RPi's clocks, we don't actually
enable/disable the firmware clock. This means that
`clk_disable_unprepare()` doesn't actually change the clock state at
all, nor does it lowers the clock rate.
From the Mailbox Property Interface documentation [1], we can see that
we should use `RPI_FIRMWARE_SET_CLOCK_STATE` to set the clock state
off/on. Therefore, use `RPI_FIRMWARE_SET_CLOCK_STATE` to create a
prepare and an unprepare hook for RPi's firmware clock.
As now the clocks are actually turned off, some of them are now marked
CLK_IS_CRITICAL, as those are required to be on during the whole system
operation.
Stefan Wahren [Thu, 31 Jul 2025 21:06:17 +0000 (18:06 -0300)]
clk: bcm: rpi: Add missing logs if firmware fails
In contrary to raspberrypi_fw_set_rate(), the ops for is_prepared() and
recalc_rate() silently ignore firmware errors by just returning 0.
Since these operations should never fail, add at least error logs
to inform the user.
Signed-off-by: Stefan Wahren <wahrenst@gmx.net> Signed-off-by: Maíra Canal <mcanal@igalia.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Jacky Bai [Mon, 28 Jul 2025 07:04:46 +0000 (15:04 +0800)]
clk: scmi: Add duty cycle ops only when duty cycle is supported
For some of the SCMI based platforms, the oem extended config may be
supported, but not for duty cycle purpose. Skip the duty cycle ops if
err return when trying to get duty cycle info.
Signed-off-by: Jacky Bai <ping.bai@nxp.com> Reviewed-by: Sudeep Holla <sudeep.holla@arm.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
The mt8183-mfgcfg node uses a power domain in its device tree node.
To prevent schema validation warnings, add the optional `power-domains`
property to the binding schema for mediatek syscon clocks.
Fixes: 1781f2c46180 ("arm64: dts: mediatek: mt8183: Add power-domains properity to mfgcfg") Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Signed-off-by: Julien Massot <julien.massot@collabora.com> Acked-by: Rob Herring (Arm) <robh@kernel.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
As described in AM335x Errata Advisory 1.0.42, WKUP_DEBUGSS_CLKCTRL
can't be disabled - the clock module will just be stuck in transitioning
state forever, resulting in the following warning message after the wait
loop times out:
l3-aon-clkctrl:0000:0: failed to disable
Just add the clock to enable_init_clks, so no attempt is made to disable
it.
Signed-off-by: Matthias Schiffer <matthias.schiffer@tq-group.com> Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com> Acked-by: Kevin Hilman <khilman@baylibre.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Marek Szyprowski [Thu, 18 Sep 2025 16:06:01 +0000 (18:06 +0200)]
clk: amlogic: fix recent code refactoring
Commit 4c4e17f27013 ("clk: amlogic: naming consistency alignment")
refactored some internals in the g12a meson clock driver. Unfortunately
it introduced a bug in the clock init data, which results in the
following kernel panic:
Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
Mem abort info:
...
Data abort info:
...
[0000000000000000] user address but active_mm is swapper
Internal error: Oops: 0000000096000004 [#1] SMP
Modules linked in:
CPU: 4 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.17.0-rc1+ #11158 PREEMPT
Hardware name: Hardkernel ODROID-N2 (DT)
pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : __clk_register+0x60/0x92c
lr : __clk_register+0x48/0x92c
...
Call trace:
__clk_register+0x60/0x92c (P)
devm_clk_hw_register+0x5c/0xd8
meson_eeclkc_probe+0x74/0x110
g12a_clkc_probe+0x2c/0x58
platform_probe+0x5c/0xac
really_probe+0xbc/0x298
__driver_probe_device+0x78/0x12c
driver_probe_device+0xdc/0x164
__driver_attach+0x9c/0x1ac
bus_for_each_dev+0x74/0xd0
driver_attach+0x24/0x30
bus_add_driver+0xe4/0x208
driver_register+0x60/0x128
__platform_driver_register+0x24/0x30
g12a_clkc_driver_init+0x1c/0x28
do_one_initcall+0x64/0x308
kernel_init_freeable+0x27c/0x4f8
kernel_init+0x20/0x1d8
ret_from_fork+0x10/0x20
Code: 52800038aa0003fcb901001852819801 (f9400260)
---[ end trace 0000000000000000 ]---
Fix this by correcting the clock init data.
Fixes: 4c4e17f27013 ("clk: amlogic: naming consistency alignment") Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on BananPi M2S Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org> Reviewed-by: Jerome Brunet <jbrunet@baylibre.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Stephen Boyd [Sat, 20 Sep 2025 04:50:38 +0000 (21:50 -0700)]
Merge tag 'sunxi-clk-for-6.18' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into clk-allwinner
Pull Allwinner clk driver updates from Chen-Yu Tsai:
In this cycle support for power-of-two single divider clocks was added.
This covers some of the clocks found in the A523 MCU PRCM clock and
reset controller, for which support was added as well.
Besides the new controller, a missing clock was added for the A523's
main clock controller. The RTC clock driver gained specifics for the
A523's RTC block for tweaking the clock rate of the internal oscillator
to get it closer to what the RTC needs.
* tag 'sunxi-clk-for-6.18' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
clk: sunxi-ng: add support for the A523/T527 MCU CCU
clk: sunxi-ng: div: support power-of-two dividers
clk: sunxi-ng: sun55i-a523-ccu: Add missing NPU module clock
dt-bindings: clock: sun55i-a523-ccu: Add A523 MCU CCU clock controller
dt-bindings: clock: sun55i-a523-ccu: Add missing NPU module clock
clk: sunxi-ng: sun6i-rtc: Add A523 specifics
Cristian Birsan [Thu, 21 Nov 2024 18:16:38 +0000 (20:16 +0200)]
ARM: at91: remove default values for PMC_PLL_ACR
Remove default values for PMC PLL Analog Control Register(ACR) as the
values are specific for each SoC and PLL and load them from PLL
characteristics structure
Co-developed-by: Andrei Simion <andrei.simion@microchip.com> Signed-off-by: Andrei Simion <andrei.simion@microchip.com> Signed-off-by: Cristian Birsan <cristian.birsan@microchip.com>
[nicolas.ferre@microchip.com: fix pll acr write sequence, preserve val] Signed-off-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Ryan Wanner [Mon, 8 Sep 2025 20:07:17 +0000 (13:07 -0700)]
clk: at91: clk-master: Add check for divide by 3
A potential divider for the master clock is div/3. The register
configuration for div/3 is MASTER_PRES_MAX. The current bit shifting
method does not work for this case. Checking for MASTER_PRES_MAX will
ensure the correct decimal value is stored in the system.
Signed-off-by: Ryan Wanner <Ryan.Wanner@microchip.com> Signed-off-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Nicolas Ferre [Wed, 27 Aug 2025 15:08:10 +0000 (17:08 +0200)]
clk: at91: clk-sam9x60-pll: force write to PLL_UPDT register
This register is important for sequencing the commands to PLLs, so
actually write the update bits with regmap_write_bits() instead of
relying on a read/modify/write regmap command that could skip the actual
hardware write if the value is identical to the one read.
It's changed when modification is needed to the PLL, when
read-only operation is done, we could keep the call to
regmap_update_bits().
Add a comment to the sam9x60_div_pll_set_div() function that uses this
PLL_UPDT register so that it's used consistently, according to the
product's datasheet.
Signed-off-by: Nicolas Ferre <nicolas.ferre@microchip.com> Tested-by: Ryan Wanner <ryan.wanner@microchip.com> # on sama7d65 and sam9x75 Link: https://lore.kernel.org/r/20250827150811.82496-1-nicolas.ferre@microchip.com
[claudiu.beznea: fix "Alignment should match open parenthesis"
checkpatch.pl check] Signed-off-by: Claudiu Beznea <claudiu.beznea@tuxon.dev>
Stephen Boyd [Wed, 17 Sep 2025 04:40:54 +0000 (21:40 -0700)]
Merge tag 'clk-imx-6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa/linux into clk-imx
Pull i.MX clk driver updates from Abel Vesa:
- Rework the i.MX95 BLK CTL driver to add the platform data to
the state container
- Retain the state of the i.MS95 BLK CTL registers through both
runtime and system suspend
* tag 'clk-imx-6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa/linux:
clk: imx95-blk-ctl: Save/restore registers when RPM routines are called
clk: imx95-blk-ctl: Save platform data in imx95_blk_ctl structure
Add "clock-output-names" which is a standard property for clock
providers.
Add the "always-on" boolean property which was undocumented, but
already in use for some time. The flag prevents a clock output from
being disabled.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org> Tested-by: Michal Simek <michal.simek@amd.com> Reviewed-by: Michal Simek <michal.simek@amd.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
* tag 'clk-meson-v6.18-1' of https://github.com/BayLibre/clk-meson:
clk: amlogic: c3-peripherals: use helper for basic composite clocks
clk: amlogic: align s4 and c3 pwm clock descriptions
clk: amlogic: add composite clock helpers
clk: amlogic: use the common pclk definition
clk: amlogic: introduce a common pclk definition
clk: amlogic: pclk explicitly use CLK_IGNORE_UNUSED
clk: amlogic: drop CLK_SET_RATE_PARENT from peripheral clocks
clk: amlogic: move PCLK definition to clkc-utils
clk: amlogic: aoclk: use clkc-utils syscon probe
clk: amlogic: use probe helper in mmio based controllers
clk: amlogic: add probe helper for mmio based controllers
clk: amlogic: drop meson-clkcee
clk: amlogic: naming consistency alignment
Stephen Boyd [Tue, 16 Sep 2025 01:11:00 +0000 (18:11 -0700)]
Merge tag 'for-6.18-clk' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux into clk-tegra
Pull Tegra clk driver updates from Thierry Reding:
- Add DFLL support on Tegra114
This is quite similar to the existing Tegra124 support and most
of the code can be reused, except for the CVB frequency tables.
* tag 'for-6.18-clk' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
clk: tegra: dfll: Add CVB tables for Tegra114
clk: tegra: Add DFLL DVCO reset control for Tegra114
dt-bindings: arm: tegra: Add ASUS TF101G and SL101
dt-bindings: reset: Add Tegra114 CAR header
dt-bindings: arm: tegra: Add Xiaomi Mi Pad (A0101)
dt-bindings: clock: tegra30: Add IDs for CSI pad clocks
dt-bindings: display: tegra: Move avdd-dsi-csi-supply from VI to CSI
dt-bindings: i2c: nvidia,tegra20-i2c: Document Tegra264 I2C
Stephen Boyd [Sat, 13 Sep 2025 22:06:14 +0000 (15:06 -0700)]
Merge tag 'samsung-clk-6.18' of https://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux into clk-samsung
Pull Samsung SoC clk driver updates from Krzysztof Kozlowski:
- Tesla FSD: Expose CSI clocks to consumers (DTS)
- Exynos990:
- Few fixes for fixed factor clocks, register widths and proper PLL
parents
- Add four more clocks for the DPU and HSI0 clock for USB
- Add PERIC0 and PERIC1 clock controllers (CMU), responsible for
providing clocks to serial engines
- Add seven clock controllers for the new Axis ARTPEC-8 SoC. The SoC
shares all main blocks, including the clock controllers, with Samsung
SoC, so same drivers and bindings are used.
- Cleanups: switch to clk_ops::determine_rate()
* tag 'samsung-clk-6.18' of https://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux:
clk: samsung: exynos990: Add PERIC0 and PERIC1 clock support
dt-bindings: clock: exynos990: Add PERIC0 and PERIC1 clock units
clk: samsung: exynos990: Add missing USB clock registers to HSI0
clk: samsung: exynos990: Add LHS_ACEL gate clock for HSI0 and update CLK_NR_TOP
dt-bindings: clock: exynos990: Add LHS_ACEL clock ID for HSI0 block
clk: samsung: artpec-8: Add initial clock support for ARTPEC-8 SoC
clk: samsung: Add clock PLL support for ARTPEC-8 SoC
dt-bindings: clock: Add ARTPEC-8 clock controller
clk: samsung: exynos990: Add DPU_BUS and CMUREF mux/div and update CLKS_NR_TOP
dt-bindings: clock: exynos990: Extend clocks IDs
clk: samsung: exynos990: Replace bogus divs with fixed-factor clocks
clk: samsung: exynos990: Fix CMU_TOP mux/div bit widths
clk: samsung: exynos990: Use PLL_CON0 for PLL parent muxes
clk: samsung: pll: convert from round_rate() to determine_rate()
clk: samsung: cpu: convert from round_rate() to determine_rate()
clk: samsung: fsd: Add clk id for PCLK and PLL in CAM_CSI block
dt-bindings: clock: Add CAM_CSI clock macro for FSD
Stephen Boyd [Sat, 13 Sep 2025 22:04:11 +0000 (15:04 -0700)]
Merge tag 'spacemit-clk-for-6.18-1' of https://github.com/spacemit-com/linux into clk-spacemit
Pull RISC-V SpacemiT clk driver updates from Yixun Lan:
- Convert to use clk_ops::determine_rate()
- Fix parent clocks of SSPA in SpacemiT driver
* tag 'spacemit-clk-for-6.18-1' of https://github.com/spacemit-com/linux:
clk: spacemit: ccu_pll: convert from round_rate() to determine_rate()
clk: spacemit: ccu_mix: convert from round_rate() to determine_rate()
clk: spacemit: ccu_ddn: convert from round_rate() to determine_rate()
clk: spacemit: fix sspax_clk
dt-bindings: clock: spacemit: CLK_SSPA_I2S_BCLK for SSPA
Stephen Boyd [Sat, 13 Sep 2025 21:56:54 +0000 (14:56 -0700)]
Merge tag 'thead-clk-for-v6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/fustini/linux into clk-thead
Pull T-HEAD clock driver updates from Drew Fustini:
- Describe gate clocks with clk_gate so that clock gates can be clock
parents. This is similar to the mux clock refactor in 54edba916e29
("clk: thead: th1520-ap: Describe mux clocks with clk_mux").
- Add support for enabling/disabling PLLs. Some PLLs are put into a
disabled state by the bootloader, and clock driver now has the
ability to enable them.
- Set all AXI clocks to CLK_IS_CRITICAL. The AXI crossbar of TH1520 has
no proper timeout handling, which means gating AXI clocks can easily
lead to bus timeout and hang the system. All these clock gates are
ungated by default on system reset.
- Convert all current CLK_IGNORE_UNUSED usage to CLK_IS_CRITICAL to
prevent unwanted clock gating.
- Fix parent of padctrl0 clock, fix parent of DPU pixel clocks and
support changing DPU pixel clock rate.
* tag 'thead-clk-for-v6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/fustini/linux:
clk: thead: th1520-ap: set all AXI clocks to CLK_IS_CRITICAL
clk: thead: support changing DPU pixel clock rate
clk: thead: add support for enabling/disabling PLLs
clk: thead: Correct parent for DPU pixel clocks
clk: thead: th1520-ap: fix parent of padctrl0 clock
clk: thead: th1520-ap: describe gate clocks with clk_gate
Stephen Boyd [Sat, 13 Sep 2025 21:33:10 +0000 (14:33 -0700)]
Merge tag 'renesas-clk-for-v6.18-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers into clk-renesas
Pull Renesas clk driver updates from Geert Uytterhoeven:
- Add Ethernet clocks on Renesas RZ/T2H and RZ/N2H
- Add USB3.0 clocks and resets on Renesas RZ/G3E
- Add I3C clocks and resets on Renesas RZ/V2H and RZ/V2N
- Add USB and remaining serial (SCI) clocks and resets on Renesas
RZ/T2H and RZ/N2H
- Add I3C and PCIe clocks and resets on Renesas RZ/G3S
- Add DMAC and PWM (GPT) clocks and resets on Renesas RZ/G3E
- Add Module Stop (MSTOP) support on RZ/G2L and Renesas RZ/G2UL
- Convert from clk_ops::round_rate() to clk_ops::determine_rate()
* tag 'renesas-clk-for-v6.18-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers: (27 commits)
clk: renesas: r9a09g05[67]: Reduce differences
clk: renesas: r9a09g047: Add USB3.0 clocks/resets
clk: renesas: cpg-mssr: Fix memory leak in cpg_mssr_reserved_init()
clk: renesas: r9a09g056: Add clock and reset entries for I3C
clk: renesas: r9a09g057: Add clock and reset entries for I3C
dt-bindings: clock: renesas,r9a09g047-cpg: Add USB3.0 core clocks
clk: renesas: r9a09g077: Add Ethernet Subsystem core and module clocks
clk: renesas: rzv2h: Simplify polling condition in __rzv2h_cpg_assert()
clk: renesas: rzv2h: Re-assert reset on deassert timeout
clk: renesas: rzg2l: Re-assert reset on deassert timeout
clk: renesas: rzg2l: Simplify rzg2l_cpg_assert() and rzg2l_cpg_deassert()
dt-bindings: clock: renesas,r9a09g077/87: Add Ethernet clock IDs
clk: renesas: r9a09g047: Add GPT clocks and resets
clk: renesas: r9a09g077: Add module clocks for SCI1-SCI5
clk: renesas: rzv2h: remove round_rate() in favor of determine_rate()
clk: renesas: rzg2l: convert from round_rate() to determine_rate()
clk: renesas: r9a07g04[34]: Use tabs instead of spaces
clk: renesas: r9a07g043: Add MSTOP for RZ/G2UL
clk: renesas: r9a07g044: Add MSTOP for RZ/G2L
clk: renesas: r9a08g045: Add MSTOP for GPIO
...
clk: sunxi-ng: add support for the A523/T527 MCU CCU
The A523/T527 SoCs have a new MCU PRCM, which has more clocks and reset
controls for the RISC-V MCU and other peripherals. There is a second
audio PLL, but no bus clock dividers. The BSP driver uses the 24MHz main
oscillator as the parent for all the bus clocks. But the diagram
suggests busses from the other PRCM are used in this block as well.
Add a driver to support this part. Unlike the BSP driver, the SoC's main
MBUS clock is chosen as the parent for the MCU MBUS clock, and the
latter then serves as the parent of the MCU DMA controller's MBUS clock.
The bus gate clocks also use their respective bus clocks as parents
according to the system bus tree diagram. In cases where a block does
not appear in that diagram, an educated guess is made.
The main clock controller on the A523/T527 has the NPU's module clock.
It was missing from the original submission, likely because that was
based on the A523 user manual; the A523 is marketed without the NPU.
Also, merge the private header back into the driver code itself. The
header only contains a macro containing the total number of clocks.
This has to be updated every time a missing clock gets added. Having
it in a separate file doesn't help the process. Instead just drop the
macro, and thus the header no longer has any reason to exist.
Also move the .num value to after the list of clks to make it obvious
that it should be updated when new clks are added.
dt-bindings: clock: sun55i-a523-ccu: Add A523 MCU CCU clock controller
There are four clock controllers in the A523 SoC. The existing binding
already covers two of them that are critical for basic operation. The
remaining ones are the MCU clock controller and CPU PLL clock
controller.
Add a description for the MCU CCU. This unit controls and provides
clocks to the MCU (RISC-V) subsystem and peripherals meant to operate
under low power conditions.
The main clock controller on the A523/T527 has the NPU's module clock.
It was missing from the original submission, likely because that was
based on the A523 user manual; the A523 is marketed without the NPU.
Laurentiu Palcu [Mon, 4 Aug 2025 13:14:50 +0000 (16:14 +0300)]
clk: imx95-blk-ctl: Save/restore registers when RPM routines are called
When runtime PM is used for clock providers that are part of a power
domain, the power domain supply is cut off during runtime suspend. This
causes all BLK CTL registers belonging to that power domain to reset. To
prevent this, save the state of the registers before entering suspend
and restore them on resume. Additionally, disable the APB clock during
suspend to minimize power consumption.
Laurentiu Palcu [Mon, 4 Aug 2025 13:14:49 +0000 (16:14 +0300)]
clk: imx95-blk-ctl: Save platform data in imx95_blk_ctl structure
Add a platform data (pdata) member to struct imx95_blk_ctl to store the
result of of_device_get_match_data() during probe to avoid redundant
calls in suspend and resume functions.
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@oss.nxp.com> Reviewed-by: Frank Li <Frank.Li@nxp.com> Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com> Reviewed-by: Abel Vesa <abel.vesa@linaro.org> Link: https://lore.kernel.org/r/20250804131450.3918846-2-laurentiu.palcu@oss.nxp.com Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
The clock drivers for RZ/V2H and RZ/V2N are very similar.
Reduce the differences between them by:
- Moving and reformatting the PLLCM33_GEAR clock definitions,
- Replacing spaces by TABs.
clk: renesas: cpg-mssr: Fix memory leak in cpg_mssr_reserved_init()
In case of krealloc_array() failure, the current error handling just
returns from the function without freeing the original array.
Fix this memory leak by freeing the original array.
Johan Hovold [Wed, 10 Sep 2025 13:47:37 +0000 (15:47 +0200)]
clk: qcom: gcc-sc8280xp: drop obsolete PCIe GDSC comment
Drop an obsolete comment about keeping the PCIe GDSCs always-on,
something which is no longer the case since commit db382dd55bcb ("clk:
qcom: gcc-sc8280xp: Allow PCIe GDSCs to enter retention state").
Abel Vesa [Wed, 30 Jul 2025 16:11:12 +0000 (19:11 +0300)]
clk: qcom: tcsrcc-x1e80100: Set the bi_tcxo as parent to eDP refclk
All the other ref clocks provided by this driver have the bi_tcxo
as parent. The eDP refclk is the only one without a parent, leading
to reporting its rate as 0. So set its parent to bi_tcxo, just like
the rest of the refclks.
The A523's RTC block is backward compatible with the R329's, but it also
has a calibration function for its internal oscillator, which would
allow it to provide a clock rate closer to the desired 32.768 KHz. This
is useful on the Radxa Cubie A5E, which does not have an external 32.768
KHz crystal.
Svyatoslav Ryhel [Fri, 29 Aug 2025 12:22:32 +0000 (15:22 +0300)]
clk: tegra: Add DFLL DVCO reset control for Tegra114
The DVCO present in the DFLL IP block has a separate reset line, exposed
via the CAR IP block. This reset line is asserted upon SoC reset.
Unless something (such as the DFLL driver) deasserts this line, the DVCO
will not oscillate, although reads and writes to the DFLL IP block will
complete.
Based on a3c83ff2 ("clk: tegra: Add DFLL DVCO reset control for Tegra124")
Svyatoslav Ryhel [Fri, 29 Aug 2025 12:22:31 +0000 (15:22 +0300)]
dt-bindings: reset: Add Tegra114 CAR header
The way that resets are handled on these Tegra devices is that there is a
set of peripheral clocks & resets which are paired up. This is because they
are laid out in banks within the CAR (clock and reset) controller. In most
cases we're referring to those resets, so you'll often see a clock ID used
in conjection with the same reset ID for a given IP block.
In addition to those peripheral resets, there are a number of extra resets
that don't have a corresponding clock and which are exposed in registers
outside of the peripheral banks, but still part of the CAR. To support
those "special" registers, the TEGRA*_RESET() is used to denote resets
outside of the regular peripheral resets. Essentially it defines the offset
within the CAR at which special resets start. In the above case, Tegra114
has 5 banks with 32 peripheral resets each. The first special reset,
TEGRA114_RESET(0), therefore gets ID 5 * 32 + 0 = 160.
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com> Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com> Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by: Thierry Reding <treding@nvidia.com>
dt-bindings: clock: tegra30: Add IDs for CSI pad clocks
Tegra30 has CSI pad clock enable bits embedded into PLLD/PLLD2 registers.
Add ids for these clocks. Additionally, move TEGRA30_CLK_CLK_MAX into
clk-tegra30 source.
dt-bindings: display: tegra: Move avdd-dsi-csi-supply from VI to CSI
The avdd-dsi-csi-supply is CSI power supply, it has nothing to do with
VI, like same supply is used with DSI and has nothing to do with DC.
Move it to correct place.
Tegra264 has 17 generic I2C controllers, two of which are in always-on
partition of the SoC. In addition to the features supported by Tegra194
it also supports a SW mutex register to allow sharing the same I2C
instance across multiple firmware.
Document compatible string "nvidia,tegra264-i2c" for Tegra264 I2C.
Jerome Brunet [Mon, 25 Aug 2025 14:26:35 +0000 (16:26 +0200)]
clk: amlogic: add composite clock helpers
Device composite clocks tend to reproduce the usual sel/div/gate
arrangement.
Add macros to help define simple composite clocks in the system.
The idea is _not_ to replace all instances of mux, div or gate with those
macros. It is rather to use it for recurring and/or simple composite
clocks, reducing controller verbosity where it makes sense. This should
help reviews focus on the tricky parts.
Jerome Brunet [Mon, 25 Aug 2025 14:26:32 +0000 (16:26 +0200)]
clk: amlogic: pclk explicitly use CLK_IGNORE_UNUSED
Every usage of CLK_IGNORE_UNUSED should be explicitly motivated and
documented. However, the PCLK macros used by most Amlogic platforms are
adding that flag systematically. Because of this, all pclks are marked with
CLK_IGNORE_UNUSED, without any form of distinction or motivation.
This may have been fine in the early days of CCF but it is not anymore.
Just removing the flag is not an option at this stage since it could cause
regression on existing platforms.
Instead, drop the flag from the macro definition and add it to the each
clock definition, for the existing clocks. This makes quite a nasty change
but it will make it a lot easier for people to contribute to fixing the
problem, clock by clock. It will also prevent new platform from being added
with a silent use of the flag.
Jerome Brunet [Mon, 25 Aug 2025 14:26:31 +0000 (16:26 +0200)]
clk: amlogic: drop CLK_SET_RATE_PARENT from peripheral clocks
On Amlogic SoCs, the rate of a peripheral clock should not be changed,
let alone the rate of the parent PLL.
These clocks are meant to be used as provided by the parent PLL. Changing
the rate would be dangerous and would likely break a lot of devices running
from the same PLL.
Don't propagate any rate change request that may come from these clocks and
drop the corresponding flag.
Jerome Brunet [Mon, 25 Aug 2025 14:26:30 +0000 (16:26 +0200)]
clk: amlogic: move PCLK definition to clkc-utils
clk-regmap was always meant to stay generic, without any amlogic specifics.
The hope was that it could move out of the amlogic directory one day.
Even if this may actually not become true, it should remain generic.
Move the amlogic peripheral clock definition out of clk-regmap header.
Jerome Brunet [Mon, 25 Aug 2025 14:26:28 +0000 (16:26 +0200)]
clk: amlogic: use probe helper in mmio based controllers
Factorize the probe function of the mmio based amlogic clock controllers
using the newly introduced probe helper. This removes a fair amount
of duplicated code.
Jerome Brunet [Mon, 25 Aug 2025 14:26:26 +0000 (16:26 +0200)]
clk: amlogic: drop meson-clkcee
What is being done by the Amlogic clock controller registration helper for
EE controllers could benefit other controllers. As such, having a specific
module for this makes little sense.
Move the helper function to clkc-utils and rename it to describe what it
does, registering syscon based controller, instead of what it serves.
Denzeel Oliva [Sun, 31 Aug 2025 12:13:14 +0000 (12:13 +0000)]
dt-bindings: clock: exynos990: Add LHS_ACEL clock ID for HSI0 block
Add the missing LHS_ACEL clock ID for the HSI0 block. This clock is
required for proper USB operation, as without it, USB connections fail
with errors like device descriptor read timeouts and address response
issues.
clk: qcom: gcc: Add support for Global Clock controller found on MSM8937
Modify existing MSM8917 driver to support MSM8937 SoC. Override frequencies
which are different in this chip. Register all the clocks to the framework
for the clients to be able to request for them. Add new variant of GDSC for
new chip.
clk: qcom: Select the intended config in QCS_DISPCC_615
Commit 9b47105f5434 ("clk: qcom: dispcc-qcs615: Add QCS615 display clock
controller driver") adds the config QCS_DISPCC_615, which selects the
non-existing config QCM_GCC_615. Probably, this is just a three-letter
abbreviation mix-up here, though. There is a config named QCS_GCC_615,
and the related config QCS_CAMCC_615 selects that config.
Fix the typo and use the intended config name in the select command.
Brian Masney [Fri, 29 Aug 2025 00:38:21 +0000 (20:38 -0400)]
clk: qcom: alpha-pll: convert from round_rate() to determine_rate()
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series.
Note that prior to running the Coccinelle,
clk_alpha_pll_postdiv_round_ro_rate() was renamed to
clk_alpha_pll_postdiv_ro_round_rate().
clk: renesas: rzv2h: Re-assert reset on deassert timeout
Prevent issues during reset deassertion by re-asserting the reset if a
timeout occurs when trying to deassert. This ensures the reset line is in a
known state and improves reliability for hardware that may not immediately
clear the reset monitor bit.