platform/kernel/linux-rpi.git
3 years agodt-bindings: clock: uniphier: Add NX1 clock binding
Kunihiko Hayashi [Tue, 12 Oct 2021 00:53:52 +0000 (09:53 +0900)]
dt-bindings: clock: uniphier: Add NX1 clock binding

Update clock binding document for UniPhier NX1 SoC.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Link: https://lore.kernel.org/r/1634000035-3114-3-git-send-email-hayashi.kunihiko@socionext.com
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: uniphier: Add audio system and video input clock control for PXs3
Kunihiko Hayashi [Tue, 12 Oct 2021 00:53:51 +0000 (09:53 +0900)]
clk: uniphier: Add audio system and video input clock control for PXs3

Add clocks for audio subsystem (AIO) and video input subsystem (EXIV) on
UniPhier PXs3 SoC.

Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Link: https://lore.kernel.org/r/1634000035-3114-2-git-send-email-hayashi.kunihiko@socionext.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: si5351: Update datasheet references
Jens Renner [Mon, 13 Sep 2021 07:48:23 +0000 (09:48 +0200)]
clk: si5351: Update datasheet references

Silicon Labs is now part of Skyworks Inc. so update the URLs to the
datasheet and application note.

Signed-off-by: Jens Renner <renner@efe-gmbh.de>
Link: https://lore.kernel.org/r/20210913074823.115212-1-renner@efe-gmbh.de
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: vc5: Use i2c .probe_new
Luca Ceresoli [Tue, 28 Sep 2021 09:50:41 +0000 (11:50 +0200)]
clk: vc5: Use i2c .probe_new

The old .probe is "soon to be deprecated". Use the new, simpler form.

Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Link: https://lore.kernel.org/r/20210928095041.17116-1-luca@lucaceresoli.net
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk/actions/owl-factor.c: remove superfluous headers
Mianhan Liu [Wed, 29 Sep 2021 06:58:24 +0000 (14:58 +0800)]
clk/actions/owl-factor.c: remove superfluous headers

owl-factor.c hasn't use any macro or function declared in linux/slab.h.
Thus, these files can be removed from owl-factor.c safely without
affecting the compilation of the ./drivers/clk module

Signed-off-by: Mianhan Liu <liumh1@shanghaitech.edu.cn>
Link: https://lore.kernel.org/r/20210929065824.23691-1-liumh1@shanghaitech.edu.cn
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: ingenic: Fix bugs with divided dividers
Paul Cercueil [Fri, 1 Oct 2021 17:20:33 +0000 (18:20 +0100)]
clk: ingenic: Fix bugs with divided dividers

Two fixes in one:

- In the "impose hardware constraints" block, the "logical" divider
  value (aka. not translated to the hardware) was clamped to fit in the
  register area, but this totally ignored the fact that the divider
  value can itself have a fixed divider.

- The code that made sure that the divider value returned by the
  function was a multiple of its own fixed divider could result in a
  wrong value being calculated, because it was rounded down instead of
  rounded up.

Fixes: 4afe2d1a6ed5 ("clk: ingenic: Allow divider value to be divided")
Co-developed-by: Artur Rojek <contact@artur-rojek.eu>
Signed-off-by: Artur Rojek <contact@artur-rojek.eu>
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Link: https://lore.kernel.org/r/20211001172033.122329-1-paul@crapouillou.net
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoMerge branches 'clk-composite-determine-fix', 'clk-allwinner', 'clk-amlogic' and...
Stephen Boyd [Tue, 2 Nov 2021 18:27:06 +0000 (11:27 -0700)]
Merge branches 'clk-composite-determine-fix', 'clk-allwinner', 'clk-amlogic' and 'clk-samsung' into clk-next

* clk-composite-determine-fix:
  clk: composite: Use rate_ops.determine_rate when also a mux is available
  clk: composite: Also consider .determine_rate for rate + mux composites

* clk-allwinner:
  clk: sunxi: sun8i-apb0: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: sun6i-ar100: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: sun6i-apb0-gates: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: sun6i-apb0: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun9i-a80-usb: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun9i-a80-de: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun9i-a80: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun8i-r40: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun8i-de2: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun8i-a83t: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun50i-h6: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun50i-a64: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: clk-mod0: Make use of the helper function devm_platform_ioremap_resource()
  dt-bindings: clocks: Fix typo in the H6 compatible
  clk: sunxi-ng: Use a separate lock for each CCU instance
  clk: sunxi-ng: Prevent unbinding CCUs via sysfs
  clk: sunxi-ng: Unregister clocks/resets when unbinding
  clk: sunxi-ng: Add machine dependency to A83T CCU
  clk: sunxi-ng: mux: Remove unused 'reg' field

* clk-amlogic:
  clk: meson: meson8b: Make the video clock trees mutable
  clk: meson: meson8b: Initialize the HDMI PLL registers
  clk: meson: meson8b: Add the HDMI PLL M/N parameters
  clk: meson: meson8b: Add the vid_pll_lvds_en gate clock
  clk: meson: meson8b: Use CLK_SET_RATE_NO_REPARENT for vclk{,2}_in_sel
  clk: meson: meson8b: Export the video clocks

* clk-samsung:
  clk: samsung: describe drivers in Kconfig
  clk: samsung: exynos5433: update apollo and atlas clock probing
  clk: samsung: add support for CPU clocks
  clk: samsung: Introduce Exynos850 clock driver
  dt-bindings: clock: Document Exynos850 CMU bindings
  dt-bindings: clock: Add bindings definitions for Exynos850 CMU
  clk: samsung: clk-pll: Implement pll0831x PLL type
  clk: samsung: clk-pll: Implement pll0822x PLL type
  clk: samsung: s5pv210-audss: Make use of devm_platform_ioremap_resource()
  clk: samsung: exynos5433: Make use of devm_platform_ioremap_resource()
  clk: samsung: exynos4412-isp: Make use of devm_platform_ioremap_resource()
  clk: samsung: exynos-audss: Make use of devm_platform_ioremap_resource()

3 years agoMerge branches 'clk-imx', 'clk-ux500' and 'clk-debugfs' into clk-next
Stephen Boyd [Tue, 2 Nov 2021 18:27:02 +0000 (11:27 -0700)]
Merge branches 'clk-imx', 'clk-ux500' and 'clk-debugfs' into clk-next

* clk-imx: (21 commits)
  clk: imx: Make CLK_IMX8ULP select MXC_CLK
  clk: imx: imx6ul: Fix csi clk gate register
  clk: imx: imx6ul: Move csi_sel mux to correct base register
  clk: imx: Fix the build break when clk-imx8ulp build as module
  clk: imx: Add the pcc reset controller support on imx8ulp
  clk: imx: Add clock driver for imx8ulp
  clk: imx: Update the pfdv2 for 8ulp specific support
  clk: imx: disable the pfd when set pfdv2 clock rate
  clk: imx: Add 'CLK_SET_RATE_NO_REPARENT' for composite-7ulp
  clk: imx: disable i.mx7ulp composite clock during initialization
  clk: imx: Update the compsite driver to support imx8ulp
  clk: imx: Update the pllv4 to support imx8ulp
  dt-bindings: clock: Add imx8ulp clock support
  clk: imx: Rework imx_clk_hw_pll14xx wrapper
  clk: imx: Rework all imx_clk_hw_composite wrappers
  clk: imx: Rework all clk_hw_register_divider wrappers
  clk: imx: Rework all clk_hw_register_mux wrappers
  clk: imx: Rework all clk_hw_register_gate2 wrappers
  clk: imx: Rework all clk_hw_register_gate wrappers
  clk: imx: Make mux/mux2 clk based helpers use clk_hw based ones
  ...

* clk-ux500:
  clk: ux500: Add driver for the reset portions of PRCC
  dt-bindings: clock: u8500: Rewrite in YAML and extend

* clk-debugfs:
  clk: use clk_core_get_rate_recalc() in clk_rate_get()

3 years agoMerge branches 'clk-leak', 'clk-rockchip', 'clk-renesas' and 'clk-at91' into clk...
Stephen Boyd [Tue, 2 Nov 2021 18:26:51 +0000 (11:26 -0700)]
Merge branches 'clk-leak', 'clk-rockchip', 'clk-renesas' and 'clk-at91' into clk-next

 - Clock power management for new SAMA7G5 SoC
 - Updates to the master clock driver and sam9x60-pll to be able to use
   cpufreq-dt driver and avoid overclocking of CPU and MCK0 domains while
   changing the frequency via DVFS
 - Power management refinement with the use of save_context()/restore_context()
   on each clock driver to specify their use in case of Backup mode only

* clk-leak:
  clk: mvebu: ap-cpu-clk: Fix a memory leak in error handling paths

* clk-rockchip:
  clk: rockchip: use module_platform_driver_probe
  clk: rockchip: rk3399: expose PCLK_COREDBG_{B,L}
  clk: rockchip: rk3399: make CPU clocks critical

* clk-renesas:
  clk: renesas: r8a779[56]x: Add MLP clocks
  clk: renesas: r9a07g044: Add SDHI clock and reset entries
  clk: renesas: rzg2l: Add SDHI clk mux support
  clk: renesas: r8a779a0: Add RPC support
  clk: renesas: cpg-lib: Move RPC clock registration to the library
  clk: renesas: r9a07g044: Add clock and reset entries for SPI Multi I/O Bus Controller
  clk: renesas: r8a779a0: Add Z0 and Z1 clock support
  clk: renesas: r9a07g044: Add GbEthernet clock/reset
  clk: renesas: rzg2l: Add support to handle coupled clocks
  clk: renesas: r9a07g044: Add ethernet clock sources
  clk: renesas: rzg2l: Add support to handle MUX clocks
  clk: renesas: r8a779a0: Add TPU clock
  clk: renesas: rzg2l: Fix clk status function
  clk: renesas: r9a07g044: Mark IA55_CLK and DMAC_ACLK critical

* clk-at91:
  clk: at91: sama7g5: set low limit for mck0 at 32KHz
  clk: at91: sama7g5: remove prescaler part of master clock
  clk: at91: clk-master: add notifier for divider
  clk: at91: clk-sam9x60-pll: add notifier for div part of PLL
  clk: at91: clk-master: fix prescaler logic
  clk: at91: clk-master: mask mckr against layout->mask
  clk: at91: clk-master: check if div or pres is zero
  clk: at91: sam9x60-pll: use DIV_ROUND_CLOSEST_ULL
  clk: at91: pmc: add sama7g5 to the list of available pmcs
  clk: at91: clk-master: improve readability by using local variables
  clk: at91: clk-master: add register definition for sama7g5's master clock
  clk: at91: sama7g5: add securam's peripheral clock
  clk: at91: pmc: execute suspend/resume only for backup mode
  clk: at91: re-factor clocks suspend/resume
  clk: at91: check pmc node status before registering syscore ops

3 years agoMerge branches 'clk-qcom', 'clk-mtk', 'clk-versatile' and 'clk-doc' into clk-next
Stephen Boyd [Tue, 2 Nov 2021 18:26:33 +0000 (11:26 -0700)]
Merge branches 'clk-qcom', 'clk-mtk', 'clk-versatile' and 'clk-doc' into clk-next

 - Use ARRAY_SIZE in qcom clk drivers
 - Remove some impractical fallback parent names in qcom clk drivers
 - GCC and RPMcc support for Qualcomm QCM2290 SoCs
 - GCC support for Qualcomm MSM8994/MSM8992 SoCs
 - LPASSCC and CAMCC support for Qualcomm SC7280 SoCs
 - Support for Mediatek MT8195 SoCs
 - Make Mediatek clk drivers tristate

* clk-qcom: (44 commits)
  clk: qcom: gdsc: enable optional power domain support
  clk: qcom: videocc-sm8250: use runtime PM for the clock controller
  clk: qcom: dispcc-sm8250: use runtime PM for the clock controller
  dt-bindings: clock: qcom,videocc: add mmcx power domain
  dt-bindings: clock: qcom,dispcc-sm8x50: add mmcx power domain
  clk: qcom: gcc-sc7280: Drop unused array
  clk: qcom: camcc: Add camera clock controller driver for SC7280
  dt-bindings: clock: Add YAML schemas for CAMCC clocks on SC7280
  clk: qcom: Add lpass clock controller driver for SC7280
  dt-bindings: clock: Add YAML schemas for LPASS clocks on SC7280
  clk: qcom: Kconfig: Sort the symbol for SC_LPASS_CORECC_7180
  clk: qcom: mmcc-sdm660: Add hw_ctrl flag to venus_core0_gdsc
  clk: qcom: mmcc-sdm660: Add necessary CXCs to venus_gdsc
  clk: qcom: gcc-msm8994: Use ARRAY_SIZE() for num_parents
  clk: qcom: gcc-msm8994: Add proper msm8992 support
  clk: qcom: gcc-msm8994: Add modem reset
  clk: qcom: gcc-msm8994: Remove the inexistent GDSC_PCIE
  clk: qcom: gcc-msm8994: Add missing clocks
  clk: qcom: gcc-msm8994: Add missing NoC clocks
  clk: qcom: gcc-msm8994: Fix up SPI QUP clocks
  ...

* clk-mtk: (28 commits)
  clk: mediatek: Export clk_ops structures to modules
  clk: mediatek: support COMMON_CLK_MT6779 module build
  clk: mediatek: support COMMON_CLK_MEDIATEK module build
  clk: composite: export clk_register_composite
  clk: mediatek: Add MT8195 apusys clock support
  clk: mediatek: Add MT8195 imp i2c wrapper clock support
  clk: mediatek: Add MT8195 wpesys clock support
  clk: mediatek: Add MT8195 vppsys1 clock support
  clk: mediatek: Add MT8195 vppsys0 clock support
  clk: mediatek: Add MT8195 vencsys clock support
  clk: mediatek: Add MT8195 vdosys1 clock support
  clk: mediatek: Add MT8195 vdosys0 clock support
  clk: mediatek: Add MT8195 vdecsys clock support
  clk: mediatek: Add MT8195 scp adsp clock support
  clk: mediatek: Add MT8195 mfgcfg clock support
  clk: mediatek: Add MT8195 ipesys clock support
  clk: mediatek: Add MT8195 imgsys clock support
  clk: mediatek: Add MT8195 ccusys clock support
  clk: mediatek: Add MT8195 camsys clock support
  clk: mediatek: Add MT8195 infrastructure clock support
  ...

* clk-versatile:
  clk: versatile: hide clock drivers from non-ARM users
  clk: versatile: Rename ICST to CLK_ICST
  clk: versatile: clk-icst: Support 'reg' in addition to 'vco-offset' for register address
  dt-bindings: clock: arm,syscon-icst: Use 'reg' instead of 'vco-offset' for VCO register address

* clk-doc:
  dt-bindings: clk: fixed-mmio-clock: Convert to YAML

3 years agoclk: use clk_core_get_rate_recalc() in clk_rate_get()
Claudiu Beznea [Mon, 11 Oct 2021 11:27:19 +0000 (14:27 +0300)]
clk: use clk_core_get_rate_recalc() in clk_rate_get()

In case clock flags contains CLK_GET_RATE_NOCACHE the clk_rate_get()
will return the cached rate. Thus, use clk_core_get_rate_recalc() which
takes proper action when clock flags contains CLK_GET_RATE_NOCACHE.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-16-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
[sboyd@kernel.org: Grab prepare lock around operation]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: sama7g5: set low limit for mck0 at 32KHz
Claudiu Beznea [Mon, 11 Oct 2021 11:27:18 +0000 (14:27 +0300)]
clk: at91: sama7g5: set low limit for mck0 at 32KHz

MCK0 could go as low as 32KHz. Set this limit.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-15-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: sama7g5: remove prescaler part of master clock
Claudiu Beznea [Mon, 11 Oct 2021 11:27:17 +0000 (14:27 +0300)]
clk: at91: sama7g5: remove prescaler part of master clock

On SAMA7G5 the prescaler part of master clock has been implemented as a
changeable one. Everytime the prescaler is changed the PMC_SR.MCKRDY bit
must be polled. Value 1 for PMC_SR.MCKRDY means the prescaler update is
done. Driver polls for this bit until it becomes 1. On SAMA7G5 it has
been discovered that in some conditions the PMC_SR.MCKRDY is not rising
but the rate it provides it's stable. The workaround is to add a timeout
when polling for PMC_SR.MCKRDY. At the moment, for SAMA7G5, the prescaler
will be removed from Linux clock tree as all the frequencies for CPU could
be obtained from PLL and also there will be less overhead when changing
frequency via DVFS.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-14-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: clk-master: add notifier for divider
Claudiu Beznea [Mon, 11 Oct 2021 11:27:16 +0000 (14:27 +0300)]
clk: at91: clk-master: add notifier for divider

SAMA7G5 supports DVFS by changing cpuck. On SAMA7G5 mck0 shares the same
parent with cpuck as seen in the following clock tree:

                       +----------> cpuck
                       |
FRAC PLL ---> DIV PLL -+-> DIV ---> mck0

mck0 could go b/w 32KHz and 200MHz on SAMA7G5. To avoid mck0 overclocking
while changing FRAC PLL or DIV PLL the commit implements a notifier for
mck0 which applies a safe divider to register (maximum value of the divider
which is 5) on PRE_RATE_CHANGE events (such that changes on PLL to not
overclock mck0) and sets the maximum allowed rate on POST_RATE_CHANGE
events.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-13-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: clk-sam9x60-pll: add notifier for div part of PLL
Claudiu Beznea [Mon, 11 Oct 2021 11:27:15 +0000 (14:27 +0300)]
clk: at91: clk-sam9x60-pll: add notifier for div part of PLL

SAM9X60's PLL which is also part of SAMA7G5 is composed of 2 parts:
one fractional part and one divider. On SAMA7G5 the CPU PLL could be
changed at run-time to implement DVFS. The hardware clock tree on
SAMA7G5 for CPU PLL is as follows:

                       +---- div1 ----------------> cpuck
                       |
FRAC PLL ---> DIV PLL -+-> prescaler ---> div0 ---> mck0

The div1 block is not implemented in Linux; on prescaler block it has
been discovered a bug on some scenarios and will be removed from Linux
in next commits. Thus, the final clock tree that will be used in Linux
will be as follows:

                       +-----------> cpuck
                       |
FRAC PLL ---> DIV PLL -+-> div0 ---> mck0

It has been proposed in [1] to not introduce a new CPUFreq driver but
to overload the proper clock drivers with proper operation such that
cpufreq-dt to be used. To accomplish this DIV PLL and div0 implement
clock notifiers which applies safe dividers before FRAC PLL is changed.
The current commit treats only the DIV PLL by adding a notifier that
sets a safe divider on PRE_RATE_CHANGE events. The safe divider is
provided by initialization clock code (sama7g5.c). The div0 is treated
in next commits (to keep the changes as clean as possible).

[1] https://lore.kernel.org/lkml/20210105104426.4tmgc2l3vyicwedd@vireshk-i7/

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-12-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: clk-master: fix prescaler logic
Claudiu Beznea [Mon, 11 Oct 2021 11:27:14 +0000 (14:27 +0300)]
clk: at91: clk-master: fix prescaler logic

When prescaler value read from register is MASTER_PRES_MAX it means
that the input clock will be divided by 3. Fix the code to reflect
this.

Fixes: 7a110b9107ed8 ("clk: at91: clk-master: re-factor master clock")
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-11-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: clk-master: mask mckr against layout->mask
Claudiu Beznea [Mon, 11 Oct 2021 11:27:13 +0000 (14:27 +0300)]
clk: at91: clk-master: mask mckr against layout->mask

Mask values read/written from/to MCKR against layout->mask as this
mask may be different b/w PMC versions.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-10-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: clk-master: check if div or pres is zero
Claudiu Beznea [Mon, 11 Oct 2021 11:27:12 +0000 (14:27 +0300)]
clk: at91: clk-master: check if div or pres is zero

Check if div or pres is zero before using it as argument for ffs().
In case div is zero ffs() will return 0 and thus substracting from
zero will lead to invalid values to be setup in registers.

Fixes: 7a110b9107ed8 ("clk: at91: clk-master: re-factor master clock")
Fixes: 75c88143f3b87 ("clk: at91: clk-master: add master clock support for SAMA7G5")
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-9-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: sam9x60-pll: use DIV_ROUND_CLOSEST_ULL
Claudiu Beznea [Mon, 11 Oct 2021 11:27:11 +0000 (14:27 +0300)]
clk: at91: sam9x60-pll: use DIV_ROUND_CLOSEST_ULL

Use DIV_ROUND_CLOSEST_ULL() to avoid any inconsistency b/w the rate
computed in sam9x60_frac_pll_recalc_rate() and the one computed in
sam9x60_frac_pll_compute_mul_frac().

Fixes: 43b1bb4a9b3e1 ("clk: at91: clk-sam9x60-pll: re-factor to support plls with multiple outputs")
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-8-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: pmc: add sama7g5 to the list of available pmcs
Claudiu Beznea [Mon, 11 Oct 2021 11:27:10 +0000 (14:27 +0300)]
clk: at91: pmc: add sama7g5 to the list of available pmcs

Add SAMA7G5 to the list of available PMCs such that the suspend/resume
code for clocks to be used on backup mode.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-7-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: clk-master: improve readability by using local variables
Claudiu Beznea [Mon, 11 Oct 2021 11:27:09 +0000 (14:27 +0300)]
clk: at91: clk-master: improve readability by using local variables

Improve readability in clk_sama7g5_master_set() by using local
variables.

Suggested-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-6-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: clk-master: add register definition for sama7g5's master clock
Claudiu Beznea [Mon, 11 Oct 2021 11:27:08 +0000 (14:27 +0300)]
clk: at91: clk-master: add register definition for sama7g5's master clock

SAMA7G5 has 4 master clocks (MCK1..4) which are controlled though the
register at offset 0x30 (relative to PMC). In the last/first phase of
suspend/resume procedure (which is architecture specific) the parent
of master clocks are changed (via assembly code) for more power saving
(see file arch/arm/mach-at91/pm_suspend.S, macros at91_mckx_ps_enable
and at91_mckx_ps_restore). Thus the macros corresponding to register
at offset 0x30 need to be shared b/w clk-master.c and pm_suspend.S.
commit ec03f18cc222 ("clk: at91: add register definition for sama7g5's
master clock") introduced the proper macros but didn't adapted the
clk-master.c as well. Thus, this commit adapt the clk-master.c to use
the macros introduced in commit ec03f18cc222 ("clk: at91: add register
definition for sama7g5's master clock").

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-5-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: sama7g5: add securam's peripheral clock
Claudiu Beznea [Mon, 11 Oct 2021 11:27:07 +0000 (14:27 +0300)]
clk: at91: sama7g5: add securam's peripheral clock

Add SECURAM's peripheral clock.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-4-claudiu.beznea@microchip.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: pmc: execute suspend/resume only for backup mode
Claudiu Beznea [Mon, 11 Oct 2021 11:27:06 +0000 (14:27 +0300)]
clk: at91: pmc: execute suspend/resume only for backup mode

Before going to backup mode architecture specific PM code sets the first
word in securam (file arch/arm/mach-at91/pm.c, function at91_pm_begin()).
Thus take this into account when suspending/resuming clocks. This will
avoid executing unnecessary instructions when suspending to non backup
modes.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-3-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: at91: re-factor clocks suspend/resume
Claudiu Beznea [Mon, 11 Oct 2021 11:27:05 +0000 (14:27 +0300)]
clk: at91: re-factor clocks suspend/resume

SAMA5D2 and SAMA7G5 have a special power saving mode (backup mode) where
most of the SoC's components are powered off (including PMC). Resuming
from this mode is done with the help of bootloader. Peripherals are not
aware of the power saving mode thus most of them are disabling clocks in
proper suspend API and re-enable them in resume API without taking into
account the previously setup rate. Moreover some of the peripherals are
acting as wakeup sources and are not disabling the clocks in this
scenario, when suspending. Since backup mode cuts the power for
peripherals, in resume part these clocks needs to be re-configured.

The initial PMC suspend/resume code was designed only for SAMA5D2's PMC
(as it was the only one supporting backup mode). SAMA7G supports also
backup mode and its PMC is different (few new functionalities, different
registers offsets, different offsets in registers for each
functionalities). To address both SAMA5D2 and SAMA7G5 PMC add
.save_context()/.resume_context() support to each clocks driver and call
this from PMC driver.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-2-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: ux500: Add driver for the reset portions of PRCC
Linus Walleij [Tue, 21 Sep 2021 18:48:03 +0000 (20:48 +0200)]
clk: ux500: Add driver for the reset portions of PRCC

The Ux500 PRCC (peripheral reset and clock controller) can also
control reset of the IP blocks, not just clocks. As the PRCC is probed
as a clock controller and we have other platforms implementing combined
clock and reset controllers, follow this pattern and implement the PRCC
rest controller as part of the clock driver.

The reset controller needs to be selected from the machine as Ux500 has
traditionally selected its mandatory subsystem prerequisites from there.

Cc: Philipp Zabel <p.zabel@pengutronix.de>
Cc: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20210921184803.1757916-2-linus.walleij@linaro.org
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
[sboyd@kernel.org: Dropped allocation error message]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clock: u8500: Rewrite in YAML and extend
Linus Walleij [Tue, 21 Sep 2021 18:48:02 +0000 (20:48 +0200)]
dt-bindings: clock: u8500: Rewrite in YAML and extend

This rewrites the ux500/u8500 clock bindings in YAML schema and extends them
with the PRCC reset controller.

The bindings are a bit idiomatic but it just reflects their age, the ux500
platform was used as guinea pig for early device tree conversion of platforms
in 2015. The new subnode for the reset controller follows the pattern of the
old bindings and adds a node with reset-cells for this.

Cc: devicetree@vger.kernel.org
Cc: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Rob Herring <robh@kernel.org>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20210921184803.1757916-1-linus.walleij@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoMerge tag 'clk-v5.16-samsung' of https://git.kernel.org/pub/scm/linux/kernel/git...
Stephen Boyd [Tue, 26 Oct 2021 19:47:45 +0000 (12:47 -0700)]
Merge tag 'clk-v5.16-samsung' of https://git./linux/kernel/git/snawrocki/clk into clk-samsung

Pull Samsung clk driver updates from Sylwester Nawrocki:

 - Initial clock driver for the Exynos850 SoC
 - Refactoring of the CPU clock code and conversion of Exynos5433
   CPU clock driver to the platform driver
 - A few conversions to devm_platform_ioremap_resource()
 - Updates of the Samsung Kconfig help text

* tag 'clk-v5.16-samsung' of https://git.kernel.org/pub/scm/linux/kernel/git/snawrocki/clk:
  clk: samsung: describe drivers in Kconfig
  clk: samsung: exynos5433: update apollo and atlas clock probing
  clk: samsung: add support for CPU clocks
  clk: samsung: Introduce Exynos850 clock driver
  dt-bindings: clock: Document Exynos850 CMU bindings
  dt-bindings: clock: Add bindings definitions for Exynos850 CMU
  clk: samsung: clk-pll: Implement pll0831x PLL type
  clk: samsung: clk-pll: Implement pll0822x PLL type
  clk: samsung: s5pv210-audss: Make use of devm_platform_ioremap_resource()
  clk: samsung: exynos5433: Make use of devm_platform_ioremap_resource()
  clk: samsung: exynos4412-isp: Make use of devm_platform_ioremap_resource()
  clk: samsung: exynos-audss: Make use of devm_platform_ioremap_resource()

3 years agoMerge tag 'clk-meson-v5.16-1' of https://github.com/BayLibre/clk-meson into clk-amlogic
Stephen Boyd [Tue, 26 Oct 2021 19:44:32 +0000 (12:44 -0700)]
Merge tag 'clk-meson-v5.16-1' of https://github.com/BayLibre/clk-meson into clk-amlogic

Pull Amlogic clock driver updates from Jerome Brunet:

 - Update video path realted clocks for meson8

* tag 'clk-meson-v5.16-1' of https://github.com/BayLibre/clk-meson:
  clk: meson: meson8b: Make the video clock trees mutable
  clk: meson: meson8b: Initialize the HDMI PLL registers
  clk: meson: meson8b: Add the HDMI PLL M/N parameters
  clk: meson: meson8b: Add the vid_pll_lvds_en gate clock
  clk: meson: meson8b: Use CLK_SET_RATE_NO_REPARENT for vclk{,2}_in_sel
  clk: meson: meson8b: Export the video clocks

3 years agoMerge tag 'sunxi-clk-for-5.16-1' of https://git.kernel.org/pub/scm/linux/kernel/git...
Stephen Boyd [Tue, 26 Oct 2021 19:37:38 +0000 (12:37 -0700)]
Merge tag 'sunxi-clk-for-5.16-1' of https://git./linux/kernel/git/sunxi/linux into clk-allwinner

Pull Allwinner clk driver updates from Maxime Ripard:

Our usual PR for the Allwinner SoCs, this time improving the module
support and converting to more helpers.

* tag 'sunxi-clk-for-5.16-1' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
  clk: sunxi: sun8i-apb0: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: sun6i-ar100: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: sun6i-apb0-gates: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: sun6i-apb0: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun9i-a80-usb: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun9i-a80-de: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun9i-a80: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun8i-r40: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun8i-de2: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun8i-a83t: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun50i-h6: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi-ng: ccu-sun50i-a64: Make use of the helper function devm_platform_ioremap_resource()
  clk: sunxi: clk-mod0: Make use of the helper function devm_platform_ioremap_resource()
  dt-bindings: clocks: Fix typo in the H6 compatible
  clk: sunxi-ng: Use a separate lock for each CCU instance
  clk: sunxi-ng: Prevent unbinding CCUs via sysfs
  clk: sunxi-ng: Unregister clocks/resets when unbinding
  clk: sunxi-ng: Add machine dependency to A83T CCU
  clk: sunxi-ng: mux: Remove unused 'reg' field

3 years agoclk: composite: Use rate_ops.determine_rate when also a mux is available
Martin Blumenstingl [Sat, 16 Oct 2021 10:50:22 +0000 (12:50 +0200)]
clk: composite: Use rate_ops.determine_rate when also a mux is available

Update clk_composite_determine_rate() to use rate_ops.determine_rate
when available in combination with a mux. So far clk_divider_ops provide
both, .round_rate and .determine_rate. Removing the former would make
clk-composite fail silently for example on Rockchip platforms (which
heavily use composite clocks).
Add support for using rate_ops.determine_rate when either
rate_ops.round_rate is not available or both (.round_rate and
.determine_rate) are provided.

Suggested-by: Alex Bee <knaerzche@gmail.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20211016105022.303413-3-martin.blumenstingl@googlemail.com
Tested-by: Alex Bee <knaerzche@gmail.com>
Tested-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: composite: Also consider .determine_rate for rate + mux composites
Martin Blumenstingl [Sat, 16 Oct 2021 10:50:21 +0000 (12:50 +0200)]
clk: composite: Also consider .determine_rate for rate + mux composites

Commit 69a00fb3d69706 ("clk: divider: Implement and wire up
.determine_rate by default") switches clk_divider_ops to implement
.determine_rate by default. This breaks composite clocks with multiple
parents because clk-composite.c does not use the special handling for
mux + divider combinations anymore (that was restricted to rate clocks
which only implement .round_rate, but not .determine_rate).

Alex reports:
  This breaks lot of clocks for Rockchip which intensively uses
  composites,  i.e. those clocks will always stay at the initial parent,
  which in some cases  is the XTAL clock and I strongly guess it is the
  same for other platforms,  which use composite clocks having more than
  one parent (e.g. mediatek, ti ...)

  Example (RK3399)
  clk_sdio is set (initialized) with XTAL (24 MHz) as parent in u-boot.
  It will always stay at this parent, even if the mmc driver sets a rate
  of  200 MHz (fails, as the nature of things), which should switch it
  to   any of its possible parent PLLs defined in
  mux_pll_src_cpll_gpll_npll_ppll_upll_24m_p (see clk-rk3399.c)  - which
  never happens.

Restore the original behavior by changing the priority of the conditions
inside clk-composite.c. Now the special rate + mux case (with rate_ops
having a .round_rate - which is still the case for the default
clk_divider_ops) is preferred over rate_ops which have .determine_rate
defined (and not further considering the mux).

Fixes: 69a00fb3d69706 ("clk: divider: Implement and wire up .determine_rate by default")
Reported-by: Alex Bee <knaerzche@gmail.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20211016105022.303413-2-martin.blumenstingl@googlemail.com
Tested-by: Alex Bee <knaerzche@gmail.com>
Tested-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: samsung: describe drivers in Kconfig
Krzysztof Kozlowski [Fri, 24 Sep 2021 13:36:24 +0000 (15:36 +0200)]
clk: samsung: describe drivers in Kconfig

Describe better which driver applies to which SoC, to make configuring
kernel for Samsung SoC easier.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Link: https://lore.kernel.org/r/20210924133624.112593-1-krzysztof.kozlowski@canonical.com
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agoMerge tag 'renesas-clk-for-v5.16-tag2' of git://git.kernel.org/pub/scm/linux/kernel...
Stephen Boyd [Fri, 15 Oct 2021 21:57:57 +0000 (14:57 -0700)]
Merge tag 'renesas-clk-for-v5.16-tag2' of git://git./linux/kernel/git/geert/renesas-drivers into clk-renesas

Pull Renesas clk driver updates from Geert Uytterhoeven:

 - Add SPI Multi I/O Bus and SDHI clocks and resets on RZ/G2L
 - Add SPI Multi I/O Bus (RPC) clocks on R-Car V3U
 - Add MediaLB clocks on R-Car H3, M3-W/W+, and M3-N

* tag 'renesas-clk-for-v5.16-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers:
  clk: renesas: r8a779[56]x: Add MLP clocks
  clk: renesas: r9a07g044: Add SDHI clock and reset entries
  clk: renesas: rzg2l: Add SDHI clk mux support
  clk: renesas: r8a779a0: Add RPC support
  clk: renesas: cpg-lib: Move RPC clock registration to the library
  clk: renesas: r9a07g044: Add clock and reset entries for SPI Multi I/O Bus Controller

3 years agoclk: samsung: exynos5433: update apollo and atlas clock probing
Will McVicker [Thu, 14 Oct 2021 19:53:46 +0000 (19:53 +0000)]
clk: samsung: exynos5433: update apollo and atlas clock probing

Use the samsung common clk driver to initialize the apollo and atlas
clocks. This removes their custom init functions and uses the
samsung_cmu_register_one() instead.

Signed-off-by: Will McVicker <willmcvicker@google.com>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Link: https://lore.kernel.org/r/20211014195347.3635601-3-willmcvicker@google.com
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agoclk: samsung: add support for CPU clocks
Will McVicker [Thu, 14 Oct 2021 19:53:45 +0000 (19:53 +0000)]
clk: samsung: add support for CPU clocks

Adds 'struct samsung_cpu_clock' and corresponding CPU clock registration
function to the samsung common clk driver. This allows samsung clock
drivers to register their CPU clocks with the samsung_cmu_register_one()
API.

Currently the exynos5433 apollo and atlas clks have their own custom
init functions to handle registering their CPU clocks. With this patch
we can drop their custom CLK_OF_DECLARE functions and directly call
samsung_cmu_register_one().

Signed-off-by: Will McVicker <willmcvicker@google.com>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Link: https://lore.kernel.org/r/20211014195347.3635601-2-willmcvicker@google.com
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agoclk: samsung: Introduce Exynos850 clock driver
Sam Protsenko [Fri, 8 Oct 2021 15:43:52 +0000 (18:43 +0300)]
clk: samsung: Introduce Exynos850 clock driver

This is the initial implementation adding only basic clocks like UART,
MMC, I2C and corresponding parent clocks. Design is influenced by
Exynos5433 clock driver.

Bus clock is enabled by default (in probe function) for all CMUs except
CMU_TOP, the reasoning is as follows. By default if bus clock has no
users its "enable count" value is 0. It might be actually running if
it's already enabled in bootloader, but then in some cases it can be
disabled by mistake. For example, such case was observed when
dw_mci_probe() enabled the bus clock, then failed to do something and
disabled that bus clock on error path. After that, even the attempt to
read the 'clk_summary' file in DebugFS freezed forever, as CMU bus clock
ended up being disabled and it wasn't possible to access CMU registers
anymore.

To avoid such cases, CMU driver must increment the ref count for that
bus clock by running clk_prepare_enable(). There is already existing
'.clk_name' field in struct samsung_cmu_info, exactly for that reason.
It was added in commit 523d3de41f02 ("clk: samsung: exynos5433: Add
support for runtime PM"), with next mentioning in commit message:

  > Also for each CMU there is one special parent clock, which has to be
  > enabled all the time when any access to CMU registers is being done.

But that clock is actually only enabled in Exynos5433 clock driver right
now. So the same code is added to exynos850_cmu_probe() function,
As was described above, it might be helpful not only for PM reasons, but
also to prevent possible erroneous clock gating on error paths.

Another way to workaround that issue would be to use CLOCK_IS_CRITICAL
flag for corresponding gate clocks. But that might be not very good
design decision, as we might still want to disable that bus clock, e.g.
on PM suspend.

Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Link: https://lore.kernel.org/r/20211008154352.19519-6-semen.protsenko@linaro.org
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agodt-bindings: clock: Document Exynos850 CMU bindings
Sam Protsenko [Fri, 8 Oct 2021 15:43:51 +0000 (18:43 +0300)]
dt-bindings: clock: Document Exynos850 CMU bindings

Provide dt-schema documentation for Exynos850 SoC clock controller.

Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Link: https://lore.kernel.org/r/20211008154352.19519-5-semen.protsenko@linaro.org
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agoclk: renesas: r8a779[56]x: Add MLP clocks
Andrey Gusakov [Wed, 29 Sep 2021 21:34:32 +0000 (00:34 +0300)]
clk: renesas: r8a779[56]x: Add MLP clocks

Add clocks for MLP modules on Renesas R-Car H3 and M3-W/N SoCs.

Signed-off-by: Andrey Gusakov <andrey.gusakov@cogentembedded.com>
Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
Link: https://lore.kernel.org/r/20210929213431.5275-1-nikita.yoush@cogentembedded.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
3 years agoclk: versatile: hide clock drivers from non-ARM users
Jean Delvare [Wed, 1 Sep 2021 16:09:53 +0000 (18:09 +0200)]
clk: versatile: hide clock drivers from non-ARM users

Commit 419b3ab6987f ("clk: versatile: remove dependency on ARCH_*")
made the whole menu of ARM reference clock drivers visible on all
architectures. I can't see how this is an improvement for non-ARM
users. Unless build-testing, there is no point on presenting
ARM-only clock drivers on other architectures.

Signed-off-by: Jean Delvare <jdelvare@suse.de>
Cc: Peter Collingbourne <pcc@google.com>
Cc: Lee Jones <lee.jones@linaro.org>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Stephen Boyd <sboyd@kernel.org>
Link: https://lore.kernel.org/r/20210901180953.5bd2a994@endymion
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: versatile: Rename ICST to CLK_ICST
Jean Delvare [Wed, 1 Sep 2021 16:08:33 +0000 (18:08 +0200)]
clk: versatile: Rename ICST to CLK_ICST

For consistency, prefix the ICST config option with CLK as all other
clock source drivers have.

Signed-off-by: Jean Delvare <jdelvare@suse.de>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Michael Turquette <mturquette@baylibre.com>
Cc: Stephen Boyd <sboyd@kernel.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20210901180833.4558932d@endymion
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gdsc: enable optional power domain support
Dmitry Baryshkov [Sun, 29 Aug 2021 15:47:54 +0000 (18:47 +0300)]
clk: qcom: gdsc: enable optional power domain support

On sm8250 dispcc and videocc registers are powered up by the MMCX power
domain. Currently we use a regulator to enable this domain on demand,
however this has some consequences, as genpd code is not reentrant.

Make gdsc code also use pm_runtime calls to ensure that registers are
accessible during the gdsc_enable/gdsc_disable operations.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210829154757.784699-6-dmitry.baryshkov@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: videocc-sm8250: use runtime PM for the clock controller
Dmitry Baryshkov [Sun, 29 Aug 2021 15:47:53 +0000 (18:47 +0300)]
clk: qcom: videocc-sm8250: use runtime PM for the clock controller

On sm8250 dispcc and videocc registers are powered up by the MMCX power
domain. Use runtime PM calls to make sure that required power domain is
powered on while we access clock controller's registers.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210829154757.784699-5-dmitry.baryshkov@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: dispcc-sm8250: use runtime PM for the clock controller
Dmitry Baryshkov [Sun, 29 Aug 2021 15:47:52 +0000 (18:47 +0300)]
clk: qcom: dispcc-sm8250: use runtime PM for the clock controller

On sm8250 dispcc and videocc registers are powered up by the MMCX power
domain. Use runtime PM calls to make sure that required power domain is
powered on while we access clock controller's registers.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210829154757.784699-4-dmitry.baryshkov@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clock: qcom,videocc: add mmcx power domain
Dmitry Baryshkov [Sun, 29 Aug 2021 15:47:51 +0000 (18:47 +0300)]
dt-bindings: clock: qcom,videocc: add mmcx power domain

On sm8250 videocc requires MMCX power domain to be powered up before
clock controller's registers become available. For now sm8250 was using
external regulator driven by the power domain to describe this
relationship. Switch into specifying power-domain and required opp-state
directly.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20210829154757.784699-3-dmitry.baryshkov@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clock: qcom,dispcc-sm8x50: add mmcx power domain
Dmitry Baryshkov [Sun, 29 Aug 2021 15:47:50 +0000 (18:47 +0300)]
dt-bindings: clock: qcom,dispcc-sm8x50: add mmcx power domain

On sm8250 dispcc requires MMCX power domain to be powered up before
clock controller's registers become available. For now sm8250 was using
external regulator driven by the power domain to describe this
relationship. Switch into specifying power-domain and required opp-state
directly.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20210829154757.784699-2-dmitry.baryshkov@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-sc7280: Drop unused array
Stephen Boyd [Thu, 14 Oct 2021 19:10:23 +0000 (12:10 -0700)]
clk: qcom: gcc-sc7280: Drop unused array

After commit 3165d1e3c737 ("clk: qcom: gcc: Remove CPUSS clocks control
for SC7280") this array is unused. Remove it.

Reported-by: kernel test robot <lkp@intel.com>
Cc: Taniya Das <tdas@codeaurora.org>
Fixes: 3165d1e3c737 ("clk: qcom: gcc: Remove CPUSS clocks control for SC7280")
Link: https://lore.kernel.org/r/20211014191259.1689641-1-sboyd@kernel.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoMerge tag 'clk-imx-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa...
Stephen Boyd [Thu, 14 Oct 2021 01:17:06 +0000 (18:17 -0700)]
Merge tag 'clk-imx-5.16' of git://git./linux/kernel/git/abelvesa/linux into clk-imx

Pull i.MX clk driver changes from Abel Vesa:

 - Remove unused helpers from i.MX specific clock header
 - Rework all clk based helpers to use clk_hw based ones
 - Rework gate/mux/divider wrappers
 - Rework imx_clk_hw_composite and imx_clk_hw_pll14xx wrappers
 - Add i.MX8ULP clock driver and related bindings
 - Update i.MX pllv4 and composite clocks to support i.MX8ULP
 - Disable i.MX7ULP composite clock during initialization
 - Add CLK_SET_RATE_NO_REPARENT flag to the i.MX7ULP composite
 - Disable the pfd when set pfdv2 clock rate
 - Add support for i.MX8ULP in pfdv2
 - Add the pcc reset controller support on i.MX8ULP
 - Fix the build break when clk-imx8ulp is built as module
 - Move csi_sel mux to correct base register in i.MX6UL clock drivr
 - Fix csi clk gate register in i.MX6UL clock driver
 - Fix build bug making CLK_IMX8ULP select MXC_CLK

* tag 'clk-imx-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa/linux: (21 commits)
  clk: imx: Make CLK_IMX8ULP select MXC_CLK
  clk: imx: imx6ul: Fix csi clk gate register
  clk: imx: imx6ul: Move csi_sel mux to correct base register
  clk: imx: Fix the build break when clk-imx8ulp build as module
  clk: imx: Add the pcc reset controller support on imx8ulp
  clk: imx: Add clock driver for imx8ulp
  clk: imx: Update the pfdv2 for 8ulp specific support
  clk: imx: disable the pfd when set pfdv2 clock rate
  clk: imx: Add 'CLK_SET_RATE_NO_REPARENT' for composite-7ulp
  clk: imx: disable i.mx7ulp composite clock during initialization
  clk: imx: Update the compsite driver to support imx8ulp
  clk: imx: Update the pllv4 to support imx8ulp
  dt-bindings: clock: Add imx8ulp clock support
  clk: imx: Rework imx_clk_hw_pll14xx wrapper
  clk: imx: Rework all imx_clk_hw_composite wrappers
  clk: imx: Rework all clk_hw_register_divider wrappers
  clk: imx: Rework all clk_hw_register_mux wrappers
  clk: imx: Rework all clk_hw_register_gate2 wrappers
  clk: imx: Rework all clk_hw_register_gate wrappers
  clk: imx: Make mux/mux2 clk based helpers use clk_hw based ones
  ...

3 years agoclk: qcom: camcc: Add camera clock controller driver for SC7280
Taniya Das [Thu, 7 Oct 2021 00:43:45 +0000 (06:13 +0530)]
clk: qcom: camcc: Add camera clock controller driver for SC7280

Add support for the camera clock controller found on SC7280 based
devices.
This would allow camera drivers to probe and control their clocks.

Signed-off-by: Taniya Das <tdas@codeaurora.org>
Link: https://lore.kernel.org/r/1633567425-11953-2-git-send-email-tdas@codeaurora.org
[sboyd@kernel.org: Make some VCOs unsigned long]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clock: Add YAML schemas for CAMCC clocks on SC7280
Taniya Das [Thu, 7 Oct 2021 00:43:44 +0000 (06:13 +0530)]
dt-bindings: clock: Add YAML schemas for CAMCC clocks on SC7280

The camera clock controller clock provider have a bunch of generic
properties that are needed in a device tree. Add the CAMCC clock IDs for
camera client to request for the clocks.

Signed-off-by: Taniya Das <tdas@codeaurora.org>
Link: https://lore.kernel.org/r/1633567425-11953-1-git-send-email-tdas@codeaurora.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: Add lpass clock controller driver for SC7280
Taniya Das [Wed, 6 Oct 2021 01:40:16 +0000 (07:10 +0530)]
clk: qcom: Add lpass clock controller driver for SC7280

Add support for the lpass clock controller found on SC7280 based devices.
This would allow lpass peripheral loader drivers to control the clocks to
bring the subsystem out of reset.

Signed-off-by: Taniya Das <tdas@codeaurora.org>
Link: https://lore.kernel.org/r/1633484416-27852-3-git-send-email-tdas@codeaurora.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clock: Add YAML schemas for LPASS clocks on SC7280
Taniya Das [Wed, 6 Oct 2021 01:40:15 +0000 (07:10 +0530)]
dt-bindings: clock: Add YAML schemas for LPASS clocks on SC7280

The LPASS(Low Power Audio Subsystem) clock provider have a bunch of generic
properties that are needed in a device tree. Add the LPASS clock IDs for
LPASS PIL client to request for the clocks.

Signed-off-by: Taniya Das <tdas@codeaurora.org>
Link: https://lore.kernel.org/r/1633484416-27852-2-git-send-email-tdas@codeaurora.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: Kconfig: Sort the symbol for SC_LPASS_CORECC_7180
Taniya Das [Wed, 6 Oct 2021 01:40:14 +0000 (07:10 +0530)]
clk: qcom: Kconfig: Sort the symbol for SC_LPASS_CORECC_7180

Fix the order of the Kconfig symbol for SC_LPASS_CORECC_7180.

Signed-off-by: Taniya Das <tdas@codeaurora.org>
Link: https://lore.kernel.org/r/1633484416-27852-1-git-send-email-tdas@codeaurora.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: mmcc-sdm660: Add hw_ctrl flag to venus_core0_gdsc
AngeloGioacchino Del Regno [Fri, 8 Oct 2021 10:20:41 +0000 (12:20 +0200)]
clk: qcom: mmcc-sdm660: Add hw_ctrl flag to venus_core0_gdsc

As shown downstream[1], this GDSC supports HW trigger mode and
we're supposed to enable it in order to ensure correct operation.

[1]: https://github.com/sonyxperiadev/kernel/blob/aosp/LA.UM.6.4.r1/arch/arm/boot/dts/qcom/sdm630.dtsi#L2181

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20211008102041.268253-2-angelogioacchino.delregno@collabora.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: mmcc-sdm660: Add necessary CXCs to venus_gdsc
AngeloGioacchino Del Regno [Fri, 8 Oct 2021 10:20:40 +0000 (12:20 +0200)]
clk: qcom: mmcc-sdm660: Add necessary CXCs to venus_gdsc

As also shown on downstream dts[1], for a correct operation of the
Venus block, we have to retain MEM/PERIPH when halting the video_core,
video_axi and video_subcore0 branches: add these CXCs to the main
Venus GDSC.

[1]: https://github.com/sonyxperiadev/kernel/blob/aosp/LA.UM.6.4.r1/arch/arm/boot/dts/qcom/sdm660-vidc.dtsi#L80

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20211008102041.268253-1-angelogioacchino.delregno@collabora.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Use ARRAY_SIZE() for num_parents
Konrad Dybcio [Thu, 23 Sep 2021 16:26:42 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Use ARRAY_SIZE() for num_parents

Don't rely on the programmer to enter the name of array elements, since the
computer can compute it with much less chance of making a mistake.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-9-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Add proper msm8992 support
Konrad Dybcio [Thu, 23 Sep 2021 16:26:41 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Add proper msm8992 support

MSM8992 is a cut-down version of MSM8994, featuring
largely the same hardware.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-8-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Add modem reset
Konrad Dybcio [Thu, 23 Sep 2021 16:26:40 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Add modem reset

This will be required to support the modem.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-7-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Remove the inexistent GDSC_PCIE
Konrad Dybcio [Thu, 23 Sep 2021 16:26:39 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Remove the inexistent GDSC_PCIE

This GDSC is not present on msm8994.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-6-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Add missing clocks
Konrad Dybcio [Thu, 23 Sep 2021 16:26:38 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Add missing clocks

This should be the last "add missing clocks" commit, as to
my knowledge there are no more clocks registered within gcc.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-5-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Add missing NoC clocks
Konrad Dybcio [Thu, 23 Sep 2021 16:26:37 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Add missing NoC clocks

Add necessary NoC clocks to provide frequency sources for
relevant branch clocks.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-4-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Fix up SPI QUP clocks
Konrad Dybcio [Thu, 23 Sep 2021 16:26:36 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Fix up SPI QUP clocks

Fix up SPI QUP freq tables to account for the fact
that not every QUP can run at the same set of frequencies.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-3-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: gcc-msm8994: Modernize the driver
Konrad Dybcio [Thu, 23 Sep 2021 16:26:35 +0000 (18:26 +0200)]
clk: qcom: gcc-msm8994: Modernize the driver

Switch to the newer-style parent_data and remove the hardcoded
xo clock.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-2-konrad.dybcio@somainline.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clk: qcom: Add bindings for MSM8994 GCC driver
Konrad Dybcio [Thu, 23 Sep 2021 16:26:34 +0000 (18:26 +0200)]
dt-bindings: clk: qcom: Add bindings for MSM8994 GCC driver

Add documentation for the MSM8994 GCC driver. While at it, retire its
compatible from the old, everyone-get-in-here file.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20210923162645.23257-1-konrad.dybcio@somainline.org
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: smd-rpm: Add QCM2290 RPM clock support
Shawn Guo [Fri, 17 Sep 2021 03:04:34 +0000 (11:04 +0800)]
clk: qcom: smd-rpm: Add QCM2290 RPM clock support

Add support for RPM-managed clocks on the QCM2290 platform.

Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Link: https://lore.kernel.org/r/20210917030434.19859-4-shawn.guo@linaro.org
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clk: qcom,rpmcc: Document QCM2290 compatible
Shawn Guo [Fri, 17 Sep 2021 03:04:33 +0000 (11:04 +0800)]
dt-bindings: clk: qcom,rpmcc: Document QCM2290 compatible

Add compatible for the RPM Clock Controller on the QCM2290 SoC.

Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Link: https://lore.kernel.org/r/20210917030434.19859-3-shawn.guo@linaro.org
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: smd-rpm: Add .recalc_rate hook for clk_smd_rpm_branch_ops
Shawn Guo [Fri, 17 Sep 2021 03:04:32 +0000 (11:04 +0800)]
clk: qcom: smd-rpm: Add .recalc_rate hook for clk_smd_rpm_branch_ops

As there is a `rate` field in clk_smd_rpm, clk_smd_rpm_recalc_rate() can
be used by branch clocks to report rate as well, rather than assuming
the rate is always same as parent clock.  This assumption doesn't hold
on platforms like QCM2290, where xo_board is 38.4MHz while bi_tcxo is
19.2MHz.

To get this work, XO buffered clocks need the following updates.

- Assign a correct rate rather than the fake one which is being used to
  generate binary value for clk_smd_rpm_req interface.

- Explicitly handle the clk_smd_rpm_req interface value for XO buffered
  clocks (.rpm_res_type being QCOM_SMD_RPM_CLK_BUF_A).

Suggested-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Link: https://lore.kernel.org/r/20210917030434.19859-2-shawn.guo@linaro.org
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
[sboyd@kernel.org: Do cpu_to_le32() again to keep sparse happy]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: Add Global Clock Controller driver for QCM2290
Shawn Guo [Sun, 19 Sep 2021 02:33:08 +0000 (10:33 +0800)]
clk: qcom: Add Global Clock Controller driver for QCM2290

Add Global Clock Controller (GCC) driver for QCM2290.  This is a porting
of gcc-scuba driver from CAF msm-4.19, with GDSC support added on top.

Because the alpha_pll on the platform has a different register
layout (offsets), its own clk_alpha_pll_regs_offset[] is used in the
driver.

Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Link: https://lore.kernel.org/r/20210919023308.24498-3-shawn.guo@linaro.org
Acked-by: Rob Herring <robh@kernel.org>
[sboyd@kernel.org: Drop duplicate includes, clk.h include, module alias]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clk: qcom: Add QCM2290 Global Clock Controller bindings
Shawn Guo [Sun, 19 Sep 2021 02:33:07 +0000 (10:33 +0800)]
dt-bindings: clk: qcom: Add QCM2290 Global Clock Controller bindings

It adds device tree bindings for QCM2290 Global Clock Controller.

Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Link: https://lore.kernel.org/r/20210919023308.24498-2-shawn.guo@linaro.org
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agodt-bindings: clock: Add bindings definitions for Exynos850 CMU
Sam Protsenko [Fri, 8 Oct 2021 15:43:50 +0000 (18:43 +0300)]
dt-bindings: clock: Add bindings definitions for Exynos850 CMU

Clock controller driver is designed to have separate instances for each
particular CMU. So clock IDs in this bindings header also start from 1
for each CMU.

Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20211008154352.19519-4-semen.protsenko@linaro.org
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agoclk: samsung: clk-pll: Implement pll0831x PLL type
Sam Protsenko [Fri, 8 Oct 2021 15:43:49 +0000 (18:43 +0300)]
clk: samsung: clk-pll: Implement pll0831x PLL type

pll0831x PLL is used in Exynos850 SoC for top-level fractional PLLs. The
code was derived from very similar pll36xx type, with next differences:

1. Lock time for pll0831x is 500*P_DIV, when for pll36xx it's 3000*P_DIV
2. It's not suggested in Exynos850 TRM that S_DIV change doesn't require
   performing PLL lock procedure (which is done in pll36xx
   implementation)
3. The offset from PMS-values register to K-value register is 0x8 for
   pll0831x, when for pll36xx it's 0x4

When defining pll0831x type, CON3 register offset should be provided as
a "con" parameter of PLL() macro, like this:

    PLL(pll_0831x, 0, "fout_mmc_pll", "oscclk",
        PLL_LOCKTIME_PLL_MMC, PLL_CON3_PLL_MMC, pll0831x_26mhz_tbl),

To define PLL rates table, one can use PLL_36XX_RATE() macro, e.g.:

    PLL_36XX_RATE(26 * MHZ, 799999877, 31, 1, 0, -15124)

as it's completely appropriate for pl0831x type and there is no sense in
duplicating that.

If bit #1 (MANUAL_PLL_CTRL) is not set in CON1 register, it won't be
possible to set new rate, with next error showing in kernel log:

    Could not lock PLL fout_mmc_pll

That can happen for example if bootloader clears that bit beforehand.
PLL driver doesn't account for that, so if MANUAL_PLL_CTRL bit was
cleared, it's assumed it was done for a reason and it shouldn't be
possible to change that PLL's rate at all.

Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Link: https://lore.kernel.org/r/20211008154352.19519-3-semen.protsenko@linaro.org
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agoclk: samsung: clk-pll: Implement pll0822x PLL type
Sam Protsenko [Fri, 8 Oct 2021 15:43:48 +0000 (18:43 +0300)]
clk: samsung: clk-pll: Implement pll0822x PLL type

pll0822x PLL is used in Exynos850 SoC for top-level integer PLLs. The
code was derived from very similar pll35xx type, with next differences:

1. Lock time for pll0822x is 150*P_DIV, when for pll35xx it's 270*P_DIV
2. It's not suggested in Exynos850 TRM that S_DIV change doesn't require
   performing PLL lock procedure (which is done in pll35xx
   implementation)

When defining pll0822x type, CON3 register offset should be provided as
a "con" parameter of PLL() macro, like this:

    PLL(pll_0822x, 0, "fout_shared0_pll", "oscclk",
        PLL_LOCKTIME_PLL_SHARED0, PLL_CON3_PLL_SHARED0,
        exynos850_shared0_pll_rates),

To define PLL rates table, one can use PLL_35XX_RATE() macro, e.g.:

    PLL_35XX_RATE(26 * MHZ, 1600 * MHZ, 800, 13, 0)

as it's completely appropriate for pl0822x type and there is no sense in
duplicating that.

If bit #1 (MANUAL_PLL_CTRL) is not set in CON1 register, it won't be
possible to set new rate, with next error showing in kernel log:

    Could not lock PLL fout_shared1_pll

That can happen for example if bootloader clears that bit beforehand.
PLL driver doesn't account for that, so if MANUAL_PLL_CTRL bit was
cleared, it's assumed it was done for a reason and it shouldn't be
possible to change that PLL's rate at all.

Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Link: https://lore.kernel.org/r/20211008154352.19519-2-semen.protsenko@linaro.org
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
3 years agoclk: renesas: r9a07g044: Add SDHI clock and reset entries
Biju Das [Thu, 7 Oct 2021 11:14:34 +0000 (12:14 +0100)]
clk: renesas: r9a07g044: Add SDHI clock and reset entries

Add SDHI{0,1} mux, clock and reset entries to CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Link: https://lore.kernel.org/r/20211007111434.8665-3-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
3 years agoclk: renesas: rzg2l: Add SDHI clk mux support
Biju Das [Thu, 7 Oct 2021 11:14:33 +0000 (12:14 +0100)]
clk: renesas: rzg2l: Add SDHI clk mux support

Add SDHI clk mux support to select SDHI clock from different clock
sources.

As per HW manual, direct clock switching from 533MHz to 400MHz and
vice versa is not recommended. So added support for handling this
in mux.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Link: https://lore.kernel.org/r/20211007111434.8665-2-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
3 years agoclk: renesas: r8a779a0: Add RPC support
Wolfram Sang [Wed, 6 Oct 2021 08:58:34 +0000 (10:58 +0200)]
clk: renesas: r8a779a0: Add RPC support

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20211006085836.42155-3-wsa+renesas@sang-engineering.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
3 years agoclk: renesas: cpg-lib: Move RPC clock registration to the library
Wolfram Sang [Wed, 6 Oct 2021 08:58:33 +0000 (10:58 +0200)]
clk: renesas: cpg-lib: Move RPC clock registration to the library

We want to reuse this code for V3U soon. Because its RPCCKCR register is
at a different offset, the moved functions do not use the base register
as an argument anymore but the RPCCKCR register itself. Verified that an
Eagle board with R-Car V3M still works.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20211006085836.42155-2-wsa+renesas@sang-engineering.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
3 years agoclk: renesas: r9a07g044: Add clock and reset entries for SPI Multi I/O Bus Controller
Lad Prabhakar [Tue, 28 Sep 2021 13:01:32 +0000 (14:01 +0100)]
clk: renesas: r9a07g044: Add clock and reset entries for SPI Multi I/O Bus Controller

Add clock and reset entries for SPI Multi I/O Bus Controller.

Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20210928130132.15022-1-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
3 years agoclk: at91: check pmc node status before registering syscore ops
Clément Léger [Mon, 13 Sep 2021 08:26:33 +0000 (10:26 +0200)]
clk: at91: check pmc node status before registering syscore ops

Currently, at91 pmc driver always register the syscore_ops whatever
the status of the pmc node that has been found. When set as secure
and disabled, the pmc should not be accessed or this will generate
abort exceptions.
To avoid this, add a check on node availability before registering
the syscore operations.

Signed-off-by: Clément Léger <clement.leger@bootlin.com>
Link: https://lore.kernel.org/r/20210913082633.110168-1-clement.leger@bootlin.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Fixes: b3b02eac33ed ("clk: at91: Add sama5d2 suspend/resume")
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoMerge tag 'renesas-clk-for-v5.16-tag1' of git://git.kernel.org/pub/scm/linux/kernel...
Stephen Boyd [Fri, 8 Oct 2021 03:41:59 +0000 (20:41 -0700)]
Merge tag 'renesas-clk-for-v5.16-tag1' of git://git./linux/kernel/git/geert/renesas-drivers into clk-renesas

Pull Renesas clk driver updates from Geert Uytterhoeven:

 - Add TPU (PWM), and Z (Cortex-A76) clocks on Renesas R-Car V3U
 - Add Ethernet clocks on Renesas RZ/G2L

* tag 'renesas-clk-for-v5.16-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers:
  clk: renesas: r8a779a0: Add Z0 and Z1 clock support
  clk: renesas: r9a07g044: Add GbEthernet clock/reset
  clk: renesas: rzg2l: Add support to handle coupled clocks
  clk: renesas: r9a07g044: Add ethernet clock sources
  clk: renesas: rzg2l: Add support to handle MUX clocks
  clk: renesas: r8a779a0: Add TPU clock
  clk: renesas: rzg2l: Fix clk status function
  clk: renesas: r9a07g044: Mark IA55_CLK and DMAC_ACLK critical

3 years agoclk: qcom: gcc: Remove CPUSS clocks control for SC7280
Taniya Das [Thu, 7 Oct 2021 04:06:11 +0000 (09:36 +0530)]
clk: qcom: gcc: Remove CPUSS clocks control for SC7280

The CPUSS clocks are kept always ON and at a fixed frequency of 100MHZ
from the bootloader and no longer required to be controlled from HLOS.

Fixes: a3cc092196ef ("clk: qcom: Add Global Clock controller (GCC) driver for SC7280")
Signed-off-by: Taniya Das <tdas@codeaurora.org>
Link: https://lore.kernel.org/r/1633579571-25475-1-git-send-email-tdas@codeaurora.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: qcom: Remove redundant .owner
Kai Song [Wed, 6 Oct 2021 04:36:27 +0000 (12:36 +0800)]
clk: qcom: Remove redundant .owner

Remove .owner field if calls are used which set it automatically
Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci

Signed-off-by: Kai Song <songkai01@inspur.com>
Link: https://lore.kernel.org/r/20211006043627.5125-1-songkai01@inspur.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: imx: Make CLK_IMX8ULP select MXC_CLK
Fabio Estevam [Wed, 6 Oct 2021 19:00:08 +0000 (16:00 -0300)]
clk: imx: Make CLK_IMX8ULP select MXC_CLK

Building CLK_IMX8ULP without selecting MXC_CLK causes the following
build errors:

ld: drivers/clk/imx/clk-imx8ulp.o: in function `imx8ulp_clk_cgc2_init':
clk-imx8ulp.c:(.text+0xd0): undefined reference to `imx_ccm_lock'
ld: clk-imx8ulp.c:(.text+0x14f): undefined reference to `imx_clk_hw_pllv4'
ld: clk-imx8ulp.c:(.text+0x15a): undefined reference to `imx_ccm_lock'

Avoid this problem by making CLK_IMX8ULP select MXC_CLK.

Fixes: c43a801a5789 ("clk: imx: Add clock driver for imx8ulp")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20211006190008.1935051-1-festevam@gmail.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: imx6ul: Fix csi clk gate register
Stefan Riedmueller [Mon, 27 Sep 2021 07:28:57 +0000 (09:28 +0200)]
clk: imx: imx6ul: Fix csi clk gate register

According to the imx6ul Reference Manual the csi clk gate register is
CCM_CCGR3 (offset 0x74) bit 0/1. For the imx6ull on the other hand the
Reference Manual lists register CCM_CCGR2 (offset 0x70) bit 2/3 as the
csi clk gate which is the current setting.

Tests have shown though that the correct csi clk gate register for the
imx6ull is actually CCM_CCGR3 bit 0/1 as well. Thus set the correct
register for both platforms.

Signed-off-by: Stefan Riedmueller <s.riedmueller@phytec.de>
Tested-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210927072857.3940880-2-s.riedmueller@phytec.de
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: imx6ul: Move csi_sel mux to correct base register
Stefan Riedmueller [Mon, 27 Sep 2021 07:28:56 +0000 (09:28 +0200)]
clk: imx: imx6ul: Move csi_sel mux to correct base register

The csi_sel mux register is located in the CCM register base and not the
CCM_ANALOG register base. So move it to the correct position in code.

Otherwise changing the parent of the csi clock can lead to a complete
system failure due to the CCM_ANALOG_PLL_SYS_TOG register being falsely
modified.

Also remove the SET_RATE_PARENT flag since one possible supply for the
csi_sel mux is the system PLL which we don't want to modify.

Signed-off-by: Stefan Riedmueller <s.riedmueller@phytec.de>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210927072857.3940880-1-s.riedmueller@phytec.de
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Fix the build break when clk-imx8ulp build as module
Jacky Bai [Fri, 17 Sep 2021 06:16:29 +0000 (14:16 +0800)]
clk: imx: Fix the build break when clk-imx8ulp build as module

Export the necessary symbols to fix the build break when clk-imx8ulp
build as module

Fixes: c43a801a5789 ("clk: imx: Add clock driver for imx8ulp")
Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210917061629.3798360-1-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Add the pcc reset controller support on imx8ulp
Jacky Bai [Tue, 14 Sep 2021 06:52:08 +0000 (14:52 +0800)]
clk: imx: Add the pcc reset controller support on imx8ulp

On i.MX8ULP, for some of the PCCs, it has a peripheral SW RST bit
resides in the same registers as the clock controller. So add this
SW RST controller support alongs with the pcc clock initialization.

the reset and clock shared the same register, to avoid  accessing
the same register by reset control and clock control concurrently,
locking is necessary, so reuse the imx_ccm_lock spinlock to simplify
the code.

Suggested-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-10-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Add clock driver for imx8ulp
Jacky Bai [Tue, 14 Sep 2021 06:52:07 +0000 (14:52 +0800)]
clk: imx: Add clock driver for imx8ulp

Add clock driver for i.MX8ULP.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-9-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Update the pfdv2 for 8ulp specific support
Jacky Bai [Tue, 14 Sep 2021 06:52:06 +0000 (14:52 +0800)]
clk: imx: Update the pfdv2 for 8ulp specific support

On i.MX8ULP, the 'CLK_SET_RATE_PARENT' flag should NOT be
set and according to the laest RM, the PFD divider value range
seems will be changed in the future, so update the pfdv2 to
include the specific support for i.MX8ULP.

Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-8-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: disable the pfd when set pfdv2 clock rate
Jacky Bai [Tue, 14 Sep 2021 06:52:05 +0000 (14:52 +0800)]
clk: imx: disable the pfd when set pfdv2 clock rate

It is possible that a PFD is enabled in HW but not in SW. That
means the enable count & prepare count of the PFD clock is '0',
so the 'CLK_SET_RATE' flag can do nothing when the rate is changed
while the PFD is hw enabled. In order to safely change the pfd
rate, we can disable the PFD directly if it is hw enabled but not
used by SW end user.

Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-7-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Add 'CLK_SET_RATE_NO_REPARENT' for composite-7ulp
Jacky Bai [Tue, 14 Sep 2021 06:52:04 +0000 (14:52 +0800)]
clk: imx: Add 'CLK_SET_RATE_NO_REPARENT' for composite-7ulp

For the imx_composite-7ulp clock type, The clock parent should
be changed explicitly by end user of this clock, if the the
'CLK_SET_RATE_NO_REPARENT' flag is not set, when user want to
set a clock frequency that can NOT get from HW accurately, then
the clock's parent will be switch to another clock parent sometimes.
This is NOT what we expected and introduced some additional debug
effort, so add the 'CLK_SET_RATE_NO_REPARENT' to avoid such unexpected
result.

Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-6-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: disable i.mx7ulp composite clock during initialization
Anson Huang [Tue, 14 Sep 2021 06:52:03 +0000 (14:52 +0800)]
clk: imx: disable i.mx7ulp composite clock during initialization

i.MX7ULP peripheral clock ONLY allow parent/rate to be changed
with clock gated, however, during clock tree initialization, the
peripheral clock could be enabled by bootloader, but the prepare
count in clock tree is still zero, so clock core driver will allow
parent/rate changed even with CLK_SET_RATE_GATE/CLK_SET_PARENT_GATE
set, but the change will fail due to HW NOT allow parent/rate change
with clock enabled. It will cause clock HW status mismatch with
clock tree info and lead to function issue. Below is an example:

usdhc0's pcc clock value is 0xC5000000 during kernel boot up, it
means usdhc0 clock is enabled, its parent is APLL_PFD1. In DT file,
the usdhc0 clock settings are as below:

assigned-clocks = <&pcc2 IMX7ULP_CLK_USDHC0>;
assigned-clock-parents = <&scg1 IMX7ULP_CLK_NIC1_DIV>;

when kernel boot up, the clock tree info is as below, but the usdhc0
PCC register is still 0xC5000000, which means its parent is still
from APLL_PFD1, which is incorrect and cause usdhc0 NOT work.

nic1_clk       2        2        0   176000000          0     0  50000
    usdhc0       0        0        0   176000000          0     0  50000

After making sure the peripheral clock is disabled during clock tree
initialization, the usdhc0 is working, and this change is necessary
for all i.MX7ULP peripheral clocks.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-5-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Update the compsite driver to support imx8ulp
Jacky Bai [Tue, 14 Sep 2021 06:52:02 +0000 (14:52 +0800)]
clk: imx: Update the compsite driver to support imx8ulp

On i.MX8ULP, some peripherals have a sw_rst control resides
in the per device PCC clock control register, all others are
same as i.MX7ULP, so update the 7ulp clock composite driver to
support i.MX8ULP to maxmimize the code reuse.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-4-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Update the pllv4 to support imx8ulp
Jacky Bai [Tue, 14 Sep 2021 06:52:01 +0000 (14:52 +0800)]
clk: imx: Update the pllv4 to support imx8ulp

The PLLs used on i.MX8ULP is mostly the same as on i.MX7ULP,
except the PLL register offset is changed. Change the PLLv4
driver for code reuse on i.MX7ULP and i.MX8ULP.

Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-3-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agodt-bindings: clock: Add imx8ulp clock support
Jacky Bai [Tue, 14 Sep 2021 06:52:00 +0000 (14:52 +0800)]
dt-bindings: clock: Add imx8ulp clock support

Add the clock dt-binding file for i.MX8ULP.

For pcc node, it will also be used as a reset controller,
so add the '#reset-cells' property description and add the
pcc reset IDs.

Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20210914065208.3582128-2-ping.bai@nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
3 years agoclk: imx: Rework imx_clk_hw_pll14xx wrapper
Abel Vesa [Mon, 13 Sep 2021 08:24:50 +0000 (11:24 +0300)]
clk: imx: Rework imx_clk_hw_pll14xx wrapper

It looks much cleaner to just have a macro compared to having
a function that passes NULL as dev to the lower-level
imx_dev_clk_hw_pll14xx.

Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/1631521490-17171-9-git-send-email-abel.vesa@nxp.com
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: imx: Rework all imx_clk_hw_composite wrappers
Abel Vesa [Mon, 13 Sep 2021 08:24:49 +0000 (11:24 +0300)]
clk: imx: Rework all imx_clk_hw_composite wrappers

Rather than having multiple different macros for each different type
of imx8m_clk_hw_composite, implement them in such a way so we can
take advantage the most of the already defined simpler types. Basically,
we end up having one low-level __imx8m_clk_hw_composite function, a
wrapper to simplify the parents related arguments called
_imx8m_clk_hw_composite and then all the types can use those for each
specific case.

Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/1631521490-17171-8-git-send-email-abel.vesa@nxp.com
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: imx: Rework all clk_hw_register_divider wrappers
Abel Vesa [Mon, 13 Sep 2021 08:24:48 +0000 (11:24 +0300)]
clk: imx: Rework all clk_hw_register_divider wrappers

Instead of having multiple inline functions that were calling
clk_hw_register_divider, implement a generic low-level
__imx_clk_hw_divider and implement the rest as macros that
pass on as arguments whatever is needed in each case.

Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/1631521490-17171-7-git-send-email-abel.vesa@nxp.com
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: imx: Rework all clk_hw_register_mux wrappers
Abel Vesa [Mon, 13 Sep 2021 08:24:47 +0000 (11:24 +0300)]
clk: imx: Rework all clk_hw_register_mux wrappers

Instead of having multiple inline functions that were calling
clk_hw_register_mux, implement a generic low-level __imx_clk_hw_mux
and implement the rest as macros that pass on as arguments whatever
is needed in each case.

Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/1631521490-17171-6-git-send-email-abel.vesa@nxp.com
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
3 years agoclk: imx: Rework all clk_hw_register_gate2 wrappers
Abel Vesa [Mon, 13 Sep 2021 08:24:46 +0000 (11:24 +0300)]
clk: imx: Rework all clk_hw_register_gate2 wrappers

Instead of having multiple inline functions that were calling
clk_hw_register_gate2, implement a generic low-level __imx_clk_hw_gate2
and implement the rest as macros that pass on as arguments whatever
is needed in each case.

Signed-off-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/1631521490-17171-5-git-send-email-abel.vesa@nxp.com
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
3 years agoMerge tag 'v5.16-rockchip-clk-1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Stephen Boyd [Tue, 28 Sep 2021 22:21:01 +0000 (15:21 -0700)]
Merge tag 'v5.16-rockchip-clk-1' of git://git./linux/kernel/git/mmind/linux-rockchip into clk-rockchip

Pull Rockchip clk driver updates from Heiko Stuebner:

 - Move to use module_platform_probe
 - Enable usage of Coresight-related clocks on rk3399

* tag 'v5.16-rockchip-clk-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:
  clk: rockchip: use module_platform_driver_probe
  clk: rockchip: rk3399: expose PCLK_COREDBG_{B,L}
  clk: rockchip: rk3399: make CPU clocks critical