platform/kernel/linux-starfive.git
2 years agoMerge branch 'clk-rate-range' into clk-next
Stephen Boyd [Fri, 14 Oct 2022 20:44:44 +0000 (13:44 -0700)]
Merge branch 'clk-rate-range' into clk-next

 - Various clk rate range fixes
 - Drop clk rate range constraints on clk_put() (redux)

* clk-rate-range: (28 commits)
  clk: mediatek: clk-mux: Add .determine_rate() callback
  clk: tests: Add tests for notifiers
  clk: Update req_rate on __clk_recalc_rates()
  clk: tests: Add missing test case for ranges
  clk: qcom: clk-rcg2: Take clock boundaries into consideration for gfx3d
  clk: Introduce the clk_hw_get_rate_range function
  clk: Zero the clk_rate_request structure
  clk: Stop forwarding clk_rate_requests to the parent
  clk: Constify clk_has_parent()
  clk: Introduce clk_core_has_parent()
  clk: Switch from __clk_determine_rate to clk_core_round_rate_nolock
  clk: Add our request boundaries in clk_core_init_rate_req
  clk: Introduce clk_hw_init_rate_request()
  clk: Move clk_core_init_rate_req() from clk_core_round_rate_nolock() to its caller
  clk: Change clk_core_init_rate_req prototype
  clk: Set req_rate on reparenting
  clk: Take into account uncached clocks in clk_set_rate_range()
  clk: tests: Add some tests for orphan with multiple parents
  clk: tests: Add tests for mux with multiple parents
  clk: tests: Add tests for single parent mux
  ...

2 years agoclk: tegra: Fix Tegra PWM parent clock
Jon Hunter [Mon, 10 Oct 2022 10:00:46 +0000 (11:00 +0100)]
clk: tegra: Fix Tegra PWM parent clock

Commit 8c193f4714df ("pwm: tegra: Optimize period calculation") updated
the period calculation in the Tegra PWM driver and now returns an error
if the period requested is less than minimum period supported. This is
breaking PWM support on various Tegra platforms. For example, on the
Tegra210 Jetson Nano platform this is breaking the PWM fan support and
probing the PWM fan driver now fails ...

 pwm-fan pwm-fan: Failed to configure PWM: -22
 pwm-fan: probe of pwm-fan failed with error -22

The problem is that the default parent clock for the PWM on Tegra210 is
a 32kHz clock and is unable to support the requested PWM period.

Fix PWM support on Tegra20, Tegra30, Tegra114, Tegra124 and Tegra210 by
updating the parent clock for the PWM to be the PLL_P.

Fixes: 8c193f4714df ("pwm: tegra: Optimize period calculation")
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Tested-by: Robert Eckelmann <longnoserob@gmail.com> # TF101 T20
Tested-by: Antoni Aloy Torrens <aaloytorrens@gmail.com> # TF101 T20
Tested-by: Svyatoslav Ryhel <clamor95@gmail.com> # TF201 T30
Tested-by: Andreas Westman Dorcsak <hedmoo@yahoo.com> # TF700T T3
Link: https://lore.kernel.org/r/20221010100046.6477-1-jonathanh@nvidia.com
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: at91: fix the build with binutils 2.27
Kefeng Wang [Wed, 12 Oct 2022 03:06:35 +0000 (11:06 +0800)]
clk: at91: fix the build with binutils 2.27

There is an issue when build with older versions of binutils 2.27.0,

arch/arm/mach-at91/pm_suspend.S: Assembler messages:
arch/arm/mach-at91/pm_suspend.S:1086: Error: garbage following instruction -- `ldr tmp1,=0x00020010UL'

Use UL() macro to fix the issue in assembly file.

Fixes: 4fd36e458392 ("ARM: at91: pm: add plla disable/enable support for sam9x60")
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Link: https://lore.kernel.org/r/20221012030635.13140-1-wangkefeng.wang@huawei.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: qcom: gcc-msm8660: Drop hardcoded fixed board clocks
Linus Walleij [Thu, 13 Oct 2022 14:07:45 +0000 (16:07 +0200)]
clk: qcom: gcc-msm8660: Drop hardcoded fixed board clocks

These two clocks are now registered in the device tree as fixed clocks,
causing a regression in the driver as the clock already exists with
e.g. the name "pxo_board" as the MSM8660 GCC driver probes.

Fix this by just not hard-coding this anymore and everything works
like a charm.

Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Fixes: baecbda52933 ("ARM: dts: qcom: msm8660: fix node names for fixed clocks")
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20221013140745.7801-1-linus.walleij@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mediatek: clk-mux: Add .determine_rate() callback
AngeloGioacchino Del Regno [Tue, 11 Oct 2022 13:55:48 +0000 (15:55 +0200)]
clk: mediatek: clk-mux: Add .determine_rate() callback

Since commit 262ca38f4b6e ("clk: Stop forwarding clk_rate_requests
to the parent"), the clk_rate_request is .. as the title says, not
forwarded anymore to the parent: this produces an issue with the
MediaTek clock MUX driver during GPU DVFS on MT8195, but not on
MT8192 or others.

This is because, differently from others, like MT8192 where all of
the clocks in the MFG parents tree are of mtk_mux type, but in the
parent tree of MT8195's MFG clock, we have one mtk_mux clock and
one (clk framework generic) mux clock, like so:

names: mfg_bg3d -> mfg_ck_fast_ref -> top_mfg_core_tmp (or) mfgpll
types: mtk_gate ->      mux        ->     mtk_mux      (or) mtk_pll

To solve this issue and also keep the GPU DVFS clocks code working
as expected, wire up a .determine_rate() callback for the mtk_mux
ops; for that, the standard clk_mux_determine_rate_flags() was used
as it was possible to.

This commit was successfully tested on MT6795 Xperia M5, MT8173 Elm,
MT8192 Spherion and MT8195 Tomato; no regressions were seen.

For the sake of some more documentation about this issue here's the
trace of it:

[   12.211587] ------------[ cut here ]------------
[   12.211589] WARNING: CPU: 6 PID: 78 at drivers/clk/clk.c:1462 clk_core_init_rate_req+0x84/0x90
[   12.211593] Modules linked in: stp crct10dif_ce mtk_adsp_common llc rfkill snd_sof_xtensa_dsp
               panfrost(+) sbs_battery cros_ec_lid_angle cros_ec_sensors snd_sof_of
               cros_ec_sensors_core hid_multitouch cros_usbpd_logger snd_sof gpu_sched
               snd_sof_utils fuse ipv6
[   12.211614] CPU: 6 PID: 78 Comm: kworker/u16:2 Tainted: G        W          6.0.0-next-20221011+ #58
[   12.211616] Hardware name: Acer Tomato (rev2) board (DT)
[   12.211617] Workqueue: devfreq_wq devfreq_monitor
[   12.211620] pstate: 40400009 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[   12.211622] pc : clk_core_init_rate_req+0x84/0x90
[   12.211625] lr : clk_core_forward_rate_req+0xa4/0xe4
[   12.211627] sp : ffff80000893b8e0
[   12.211628] x29: ffff80000893b8e0 x28: ffffdddf92f9b000 x27: ffff46a2c0e8bc05
[   12.211632] x26: ffff46a2c1041200 x25: 0000000000000000 x24: 00000000173eed80
[   12.211636] x23: ffff80000893b9c0 x22: ffff80000893b940 x21: 0000000000000000
[   12.211641] x20: ffff46a2c1039f00 x19: ffff46a2c1039f00 x18: 0000000000000000
[   12.211645] x17: 0000000000000038 x16: 000000000000d904 x15: 0000000000000003
[   12.211649] x14: ffffdddf9357ce48 x13: ffffdddf935e71c8 x12: 000000000004803c
[   12.211653] x11: 00000000a867d7ad x10: 00000000a867d7ad x9 : ffffdddf90c28df4
[   12.211657] x8 : ffffdddf9357a980 x7 : 0000000000000000 x6 : 0000000000000004
[   12.211661] x5 : ffffffffffffffc8 x4 : 00000000173eed80 x3 : ffff80000893b940
[   12.211665] x2 : 00000000173eed80 x1 : ffff80000893b940 x0 : 0000000000000000
[   12.211669] Call trace:
[   12.211670]  clk_core_init_rate_req+0x84/0x90
[   12.211673]  clk_core_round_rate_nolock+0xe8/0x10c
[   12.211675]  clk_mux_determine_rate_flags+0x174/0x1f0
[   12.211677]  clk_mux_determine_rate+0x1c/0x30
[   12.211680]  clk_core_determine_round_nolock+0x74/0x130
[   12.211682]  clk_core_round_rate_nolock+0x58/0x10c
[   12.211684]  clk_core_round_rate_nolock+0xf4/0x10c
[   12.211686]  clk_core_set_rate_nolock+0x194/0x2ac
[   12.211688]  clk_set_rate+0x40/0x94
[   12.211691]  _opp_config_clk_single+0x38/0xa0
[   12.211693]  _set_opp+0x1b0/0x500
[   12.211695]  dev_pm_opp_set_rate+0x120/0x290
[   12.211697]  panfrost_devfreq_target+0x3c/0x50 [panfrost]
[   12.211705]  devfreq_set_target+0x8c/0x2d0
[   12.211707]  devfreq_update_target+0xcc/0xf4
[   12.211708]  devfreq_monitor+0x40/0x1d0
[   12.211710]  process_one_work+0x294/0x664
[   12.211712]  worker_thread+0x7c/0x45c
[   12.211713]  kthread+0x104/0x110
[   12.211716]  ret_from_fork+0x10/0x20
[   12.211718] irq event stamp: 7102
[   12.211719] hardirqs last  enabled at (7101): [<ffffdddf904ea5a0>] finish_task_switch.isra.0+0xec/0x2f0
[   12.211723] hardirqs last disabled at (7102): [<ffffdddf91794b74>] el1_dbg+0x24/0x90
[   12.211726] softirqs last  enabled at (6716): [<ffffdddf90410be4>] __do_softirq+0x414/0x588
[   12.211728] softirqs last disabled at (6507): [<ffffdddf904171d8>] ____do_softirq+0x18/0x24
[   12.211730] ---[ end trace 0000000000000000 ]---

Fixes: 262ca38f4b6e ("clk: Stop forwarding clk_rate_requests to the parent")
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20221011135548.318323-1-angelogioacchino.delregno@collabora.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: tests: Add tests for notifiers
Maxime Ripard [Mon, 10 Oct 2022 14:47:39 +0000 (16:47 +0200)]
clk: tests: Add tests for notifiers

We're recently encountered a regression due to the rates reported
through the clk_notifier_data being off when changing parents.

Let's add a test suite and a test to make sure that we do get notified
and with the proper rates.

Suggested-by: Stephen Boyd <sboyd@kernel.org>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://lore.kernel.org/r/20221010-rpi-clk-fixes-again-v1-2-d87ba82ac404@cerno.tech
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: Update req_rate on __clk_recalc_rates()
Maxime Ripard [Mon, 10 Oct 2022 14:47:38 +0000 (16:47 +0200)]
clk: Update req_rate on __clk_recalc_rates()

Commit cb1b1dd96241 ("clk: Set req_rate on reparenting") introduced a
new function, clk_core_update_orphan_child_rates(), that updates the
req_rate field on reparenting.

It turns out that that function will interfere with the clock notifying
done by __clk_recalc_rates(). This ends up reporting the new rate in
both the old_rate and new_rate fields of struct clk_notifier_data.

Since clk_core_update_orphan_child_rates() is basically
__clk_recalc_rates() without the notifiers, and with the req_rate field
update, we can drop clk_core_update_orphan_child_rates() entirely, and
make __clk_recalc_rates() update req_rate.

However, __clk_recalc_rates() is being called in several code paths:
when retrieving a rate (most likely through clk_get_rate()), when changing
parents (through clk_set_rate() or clk_hw_reparent()), or when updating
the orphan status (through clk_core_reparent_orphans_nolock(), called at
registration).

Updating req_rate on reparenting or initialisation makes sense, but we
shouldn't do it on clk_get_rate(). Thus an extra flag has been added to
update or not req_rate depending on the context.

Fixes: cb1b1dd96241 ("clk: Set req_rate on reparenting")
Link: https://lore.kernel.org/linux-clk/0acc7217-762c-7c0d-45a0-55c384824ce4@samsung.com/
Link: https://lore.kernel.org/linux-clk/Y0QNSx+ZgqKSvPOC@sirena.org.uk/
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reported-by: Mark Brown <broonie@kernel.org>
Suggested-by: Stephen Boyd <sboyd@kernel.org>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://lore.kernel.org/r/20221010-rpi-clk-fixes-again-v1-1-d87ba82ac404@cerno.tech
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoMerge branches 'clk-baikal', 'clk-broadcom', 'clk-vc5' and 'clk-versaclock' into...
Stephen Boyd [Tue, 4 Oct 2022 17:54:34 +0000 (10:54 -0700)]
Merge branches 'clk-baikal', 'clk-broadcom', 'clk-vc5' and 'clk-versaclock' into clk-next

 - Convert Baikal-T1 CCU driver to platform driver
 - Split reset support out of primary Baikal-T1 CCU driver
 - Add some missing clks required for RPiVid Video Decoder on RaspberryPi
 - Mark PLLC critical on bcm2835
 - Support for Renesas VersaClock7 clock generator family

* clk-baikal:
  clk: baikal-t1: Convert to platform device driver
  clk: baikal-t1: Add DDR/PCIe directly controlled resets support
  dt-bindings: clk: baikal-t1: Add DDR/PCIe reset IDs
  clk: baikal-t1: Move reset-controls code into a dedicated module
  clk: baikal-t1: Add SATA internal ref clock buffer
  clk: baikal-t1: Add shared xGMAC ref/ptp clocks internal parent
  clk: baikal-t1: Fix invalid xGMAC PTP clock divider
  clk: vc5: Fix 5P49V6901 outputs disabling when enabling FOD

* clk-broadcom:
  clk: bcm: rpi: Add support for VEC clock
  clk: bcm: rpi: Handle pixel clock in firmware
  clk: bcm: rpi: Add support HEVC clock
  clk: bcm2835: fix bcm2835_clock_rate_from_divisor declaration
  clk: bcm2835: Round UART input clock up
  clk: bcm2835: Make peripheral PLLC critical

* clk-vc5:
  clk: vc5: Add support for IDT/Renesas VersaClock 5P49V6975
  dt-bindings: clock: vc5: Add 5P49V6975
  clk: vc5: Use regmap_{set,clear}_bits() where appropriate
  clk: vc5: Check IO access results

* clk-versaclock:
  clk: Renesas versaclock7 ccf device driver
  dt-bindings: Renesas versaclock7 device tree bindings

2 years agoMerge branches 'clk-fixed-rate', 'clk-spreadtrum', 'clk-pxa' and 'clk-ti' into clk...
Stephen Boyd [Tue, 4 Oct 2022 17:54:14 +0000 (10:54 -0700)]
Merge branches 'clk-fixed-rate', 'clk-spreadtrum', 'clk-pxa' and 'clk-ti' into clk-next

 - More devm helpers for fixed rate registration
 - Add Spreadtrum UMS512 SoC clk support
 - Various PXA168 clk driver fixes

* clk-fixed-rate:
  clk: fixed-rate: add devm_clk_hw_register_fixed_rate
  clk: asm9260: use parent index to link the reference clock

* clk-spreadtrum:
  clk: sprd: Add clocks support for UMS512

* clk-pxa:
  clk: pxa: add a check for the return value of kzalloc()
  clk: mmp: pxa168: control shared SDH bits with separate clock
  dt-bindings: marvell,pxa168: add clock ids for SDH AXI clocks
  clk: mmp: pxa168: add clocks for SDH2 and SDH3
  dt-bindings: marvell,pxa168: add clock id for SDH3
  clk: mmp: pxa168: fix GPIO clock enable bits
  clk: mmp: pxa168: add muxes for more peripherals
  clk: mmp: pxa168: fix incorrect parent clocks
  clk: mmp: pxa168: fix const-correctness
  clk: mmp: pxa168: add new clocks for peripherals
  dt-bindings: marvell,pxa168: add clock ids for additional dividers
  clk: mmp: pxa168: fix incorrect dividers
  clk: mmp: pxa168: add additional register defines

* clk-ti:
  clk: davinci: cfgchip: Use dev_err_probe() helper
  clk: davinci: pll: fix spelling typo in comment
  MAINTAINERS: add header file to TI DAVINCI SERIES CLOCK DRIVER

2 years agoMerge branches 'clk-rockchip', 'clk-renesas', 'clk-microchip', 'clk-allwinner' and...
Stephen Boyd [Tue, 4 Oct 2022 17:54:02 +0000 (10:54 -0700)]
Merge branches 'clk-rockchip', 'clk-renesas', 'clk-microchip', 'clk-allwinner' and 'clk-imx' into clk-next

* clk-rockchip:
  dt-bindings: clock: rockchip: change SPDX-License-Identifier
  dt-bindings: clock: convert rockchip,rk3128-cru.txt to YAML
  clk: rockchip: Add clock controller support for RV1126 SoC
  dt-bindings: clock: rockchip: Document RV1126 CRU
  clk: rockchip: Add dt-binding header for RV1126
  clk: rockchip: Add MUXTBL variant

* clk-renesas:
  clk: renesas: r8a779g0: Add EtherAVB clocks
  clk: renesas: r8a779g0: Add PFC/GPIO clocks
  clk: renesas: r8a779g0: Add I2C clocks
  clk: renesas: r8a779g0: Add watchdog clock
  dt-bindings: clock: renesas,rzg2l: Document RZ/Five SoC
  clk: renesas: r8a779f0: Add MSIOF clocks
  clk: renesas: r9a09g011: Add IIC clock and reset entries
  clk: renesas: r9a07g044: Add conditional compilation for r9a07g044_cpg_info
  clk: renesas: r8a779f0: Add TMU and parent SASYNC clocks
  clk: renesas: r8a779f0: Add CMT clocks
  clk: renesas: r8a779f0: Add SDH0 clock

* clk-microchip:
  clk: at91: sama5d2: Add Generic Clocks for UART/USART
  clk: microchip: add PolarFire SoC fabric clock support
  dt-bindings: clk: add PolarFire SoC fabric clock ids
  dt-bindings: clk: document PolarFire SoC fabric clocks
  dt-bindings: clk: rename mpfs-clkcfg binding
  clk: microchip: mpfs: update module authorship & licencing
  clk: microchip: mpfs: convert periph_clk to clk_gate
  clk: microchip: mpfs: convert cfg_clk to clk_divider
  clk: microchip: mpfs: delete 2 line mpfs_clk_register_foo()
  clk: microchip: mpfs: simplify control reg access
  clk: microchip: mpfs: move id & offset out of clock structs
  clk: microchip: mpfs: add MSS pll's set & round rate
  MAINTAINERS: add polarfire soc reset controller
  reset: add polarfire soc reset support
  clk: microchip: mpfs: add reset controller
  dt-bindings: clk: microchip: mpfs: add reset controller support
  clk: microchip: mpfs: make the rtc's ahb clock critical
  clk: microchip: mpfs: fix clk_cfg array bounds violation

* clk-allwinner:
  clk: sunxi-ng: ccu-sun9i-a80-usb: Use dev_err_probe() helper
  clk: sunxi-ng: ccu-sun9i-a80-de: Use dev_err_probe() helper
  clk: sunxi-ng: sun8i-de2: Use dev_err_probe() helper
  clk: sunxi-ng: d1: Limit PLL rates to stable ranges

* clk-imx:
  clk: imx: scu: fix memleak on platform_device_add() fails
  clk: imx93: add SAI IPG clk
  clk: imx93: add MU1/2 clock
  clk: imx93: switch to use new clk gate API
  clk: imx: add i.MX93 clk gate
  clk: imx: clk-composite-93: check white_list
  clk: imx: clk-composite-93: check slice busy
  dt-bindings: clock: imx93-clock: add more MU/SAI clocks
  dt-bindings: clock: imx8mm: don't use multiple blank lines
  clk: imx8mp: tune the order of enet_qos_root_clk

2 years agoMerge branches 'clk-samsung', 'clk-mtk', 'clk-rm', 'clk-ast' and 'clk-qcom' into...
Stephen Boyd [Tue, 4 Oct 2022 17:53:41 +0000 (10:53 -0700)]
Merge branches 'clk-samsung', 'clk-mtk', 'clk-rm', 'clk-ast' and 'clk-qcom' into clk-next

 - Add resets for MediaTek MT8195 PCIe and USB
 - Remove DaVinci DM644x and DM646x clk driver support

* clk-samsung:
  clk: samsung: MAINTAINERS: add Krzysztof Kozlowski
  clk: samsung: exynos850: Implement CMU_MFCMSCL domain
  clk: samsung: exynos850: Implement CMU_IS domain
  clk: samsung: exynos850: Implement CMU_AUD domain
  clk: samsung: exynos850: Style fixes
  clk: samsung: exynosautov9: add fsys1 clock support
  clk: samsung: exynosautov9: add fsys0 clock support
  clk: samsung: exynosautov9: correct register offsets of peric0/c1
  clk: samsung: exynosautov9: add missing gate clks for peric0/c1
  dt-bindings: clock: exynos850: Add Exynos850 CMU_MFCMSCL
  dt-bindings: clock: exynos850: Add Exynos850 CMU_IS
  dt-bindings: clock: exynos850: Add Exynos850 CMU_AUD
  dt-bindings: clock: exynosautov9: add schema for cmu_fsys0/1
  dt-bindings: clock: exynosautov9: add fsys1 clock definitions
  dt-bindings: clock: exynosautov9: add fys0 clock definitions
  clk: samsung: exynos7885: Add TREX clocks
  clk: samsung: exynos7885: Implement CMU_FSYS domain
  dt-bindings: clock: exynosautov9: correct clock numbering of peric0/c1
  clk: samsung: exynos-clkout: Use of_device_get_match_data()

* clk-mtk: (42 commits)
  clk: mediatek: add driver for MT8365 SoC
  clk: mediatek: Export required common code symbols
  clk: mediatek: Provide mtk_devm_alloc_clk_data
  dt-bindings: clock: mediatek: add bindings for MT8365 SoC
  clk: mediatek: mt8192: deduplicate parent clock lists
  clk: mediatek: Migrate remaining clk_unregister_*() to clk_hw_unregister_*()
  clk: mediatek: fix unregister function in mtk_clk_register_dividers cleanup
  clk: mediatek: clk-mt8192: Add clock mux notifier for mfg_pll_sel
  clk: mediatek: clk-mt8192-mfg: Propagate rate changes to parent
  clk: mediatek: clk-mt8195-topckgen: Drop univplls from mfg mux parents
  clk: mediatek: clk-mt8195-topckgen: Add GPU clock mux notifier
  clk: mediatek: clk-mt8195-topckgen: Register mfg_ck_fast_ref as generic mux
  clk: mediatek: clk-mt8195-mfg: Reparent mfg_bg3d and propagate rate changes
  clk: mediatek: mt8183: Add clk mux notifier for MFG mux
  clk: mediatek: mux: add clk notifier functions
  clk: mediatek: mt8183: mfgcfg: Propagate rate changes to parent
  clk: mediatek: Use mtk_clk_register_gates_with_dev in simple probe
  clk: mediatek: gate: Export mtk_clk_register_gates_with_dev
  clk: mediatek: add VDOSYS1 clock
  dt-bindings: clk: mediatek: Add MT8195 DPI clocks
  ...

* clk-rm:
  clk: davinci: remove PLL and PSC clocks for DaVinci DM644x and DM646x

* clk-ast:
  clk: ast2600: BCLK comes from EPLL

* clk-qcom: (97 commits)
  clk: qcom: gcc-sm6375: Ensure unsigned long type
  clk: qcom: gcc-sm6375: Remove unused variables
  clk: qcom: kpss-xcc: convert to parent data API
  clk: introduce (devm_)hw_register_mux_parent_data_table API
  clk: qcom: gcc-msm8939: use ARRAY_SIZE instead of specifying num_parents
  clk: qcom: gcc-msm8939: use parent_hws where possible
  dt-bindings: clock: move qcom,gcc-msm8939 to qcom,gcc-msm8916.yaml
  clk: qcom: gcc-sm6350: Update the .pwrsts for usb gdscs
  clk: qcom: gcc-sc8280xp: use retention for USB power domains
  clk: qcom: gdsc: add missing error handling
  dt-bindings: clocks: qcom,gcc-sc8280xp: Fix typos
  clk: qcom: Add global clock controller driver for SM6375
  dt-bindings: clock: add SM6375 QCOM global clock bindings
  clk: qcom: alpha: Add support for programming the PLL_FSM_LEGACY_MODE bit
  clk: qcom: gcc-sc7280: Update the .pwrsts for usb gdscs
  clk: qcom: gcc-sc7180: Update the .pwrsts for usb gdsc
  clk: qcom: gdsc: Fix the handling of PWRSTS_RET support
  clk: qcom: Add SC8280XP GPU clock controller
  dt-bindings: clock: Add Qualcomm SC8280XP GPU binding
  clk: qcom: smd: Add SM6375 clocks
  ...

2 years agoMerge branches 'clk-ofnode', 'clk-bindings', 'clk-cleanup', 'clk-zynq' and 'clk-xilin...
Stephen Boyd [Tue, 4 Oct 2022 17:53:04 +0000 (10:53 -0700)]
Merge branches 'clk-ofnode', 'clk-bindings', 'clk-cleanup', 'clk-zynq' and 'clk-xilinx' into clk-next

 - Miscellaneous of_node_put() fixes
 - Nuke dt-bindings/clk path (again) by moving headers to dt-bindings/clock
 - Convert gpio-clk-gate binding to YAML
 - Various fixes to AMD/Xilinx Zynqmp clk driver
 - Graduate AMD/Xilinx "clocking wizard" driver from staging

* clk-ofnode:
  clk: ti: Balance of_node_get() calls for of_find_node_by_name()
  clk: tegra20: Fix refcount leak in tegra20_clock_init
  clk: tegra: Fix refcount leak in tegra114_clock_init
  clk: tegra: Fix refcount leak in tegra210_clock_init
  clk: sprd: Hold reference returned by of_get_parent()
  clk: berlin: Add of_node_put() for of_get_parent()
  clk: at91: dt-compat: Hold reference returned by of_get_parent()
  clk: qoriq: Hold reference returned by of_get_parent()
  clk: oxnas: Hold reference returned by of_get_parent()
  clk: st: Hold reference returned by of_get_parent()
  clk: tegra: Add missing of_node_put()
  clk: meson: Hold reference returned by of_get_parent()
  clk: nomadik: Add missing of_node_put()

* clk-bindings:
  dt-bindings: clock: drop minItems equal to maxItems
  dt-bindings: clock: gpio-gate-clock: Convert to json-schema
  dt-bindings: clock: Move versaclock.h to dt-bindings/clock
  dt-bindings: clock: Move lochnagar.h to dt-bindings/clock

* clk-cleanup:
  clk: allow building lan966x as a module
  clk: clk-xgene: simplify if-if to if-else
  clk: nxp: fix typo in comment
  clk: mvebu: armada-37xx-tbg: Remove the unneeded result variable
  clk: ti: dra7-atl: Fix reference leak in of_dra7_atl_clk_probe
  clkdev: Simplify devm_clk_hw_register_clkdev() function
  clkdev: Remove never used devm_clk_release_clkdev()
  clk: Remove never used devm_of_clk_del_provider()
  clk: pistachio: Fix initconst confusion
  clk: clk-npcm7xx: Remove unused struct npcm7xx_clk_gate_data and npcm7xx_clk_div_fixed_data
  clk: do not initialize ret
  clk: remove extra empty line
  clk: Fix comment typo
  clk: move from strlcpy with unused retval to strscpy

* clk-zynq:
  clk: zynqmp: pll: rectify rate rounding in zynqmp_pll_round_rate
  clk: zynqmp: Check the return type zynqmp_pm_query_data
  clk: zynqmp: Add a check for NULL pointer
  clk: zynqmp: Replaced strncpy() with strscpy()
  clk: zynqmp: Fix stack-out-of-bounds in strncpy`
  clk: zynqmp: make bestdiv unsigned

* clk-xilinx:
  clk: clocking-wizard: Depend on HAS_IOMEM
  clk: clocking-wizard: Use dev_err_probe() helper
  clk: clocking-wizard: Update the compatible
  clk: clocking-wizard: Fix the reconfig for 5.2
  clk: clocking-wizard: Rename nr-outputs to xlnx,nr-outputs
  clk: clocking-wizard: Move clocking-wizard out
  dt-bindings: add documentation of xilinx clocking wizard

2 years agoclk: qcom: gcc-sm6375: Ensure unsigned long type
Stephen Boyd [Tue, 4 Oct 2022 17:17:23 +0000 (10:17 -0700)]
clk: qcom: gcc-sm6375: Ensure unsigned long type

This PLL frequency needs a UL postfix to avoid compiler warnings on
32-bit architectures.

Fixes: 184fdd873d83 ("clk: qcom: Add global clock controller driver for SM6375")
Cc: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: qcom: gcc-sm6375: Remove unused variables
Konrad Dybcio [Mon, 3 Oct 2022 21:14:38 +0000 (23:14 +0200)]
clk: qcom: gcc-sm6375: Remove unused variables

gcc_parent_data_15 and gcc_parent_map_15 are not used in this driver.
Remove them.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Link: https://lore.kernel.org/r/20221003211438.25691-1-konrad.dybcio@somainline.org
Fixes: 184fdd873d83 ("clk: qcom: Add global clock controller driver for SM6375")
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: qcom: kpss-xcc: convert to parent data API
Christian Marangi [Wed, 14 Sep 2022 14:47:43 +0000 (16:47 +0200)]
clk: qcom: kpss-xcc: convert to parent data API

Convert the driver to parent data API. From the Documentation pll8_vote
and pxo should be declared in the DTS so fw_name can be used instead of
parent_names. .name is changed to the legacy pxo_board following how
it's declared in other drivers.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://lore.kernel.org/r/20220914144743.17369-2-ansuelsmth@gmail.com
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: introduce (devm_)hw_register_mux_parent_data_table API
Christian Marangi [Wed, 14 Sep 2022 14:47:42 +0000 (16:47 +0200)]
clk: introduce (devm_)hw_register_mux_parent_data_table API

Introduce (devm_)hw_register_mux_parent_data_table new API. We have
basic support for clk_register_mux using parent_data but we lack any API
to provide a custom parent_map. Add these 2 new API to correctly handle
these special configuration instead of using the generic
__(devm_)clk_hw_register_mux API.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://lore.kernel.org/r/20220914144743.17369-1-ansuelsmth@gmail.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoMerge tag 'qcom-clk-for-6.1' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom...
Stephen Boyd [Tue, 4 Oct 2022 03:48:41 +0000 (20:48 -0700)]
Merge tag 'qcom-clk-for-6.1' of https://git./linux/kernel/git/qcom/linux into clk-qcom

Pull Qualcomm clk driver updates from Bjorn Andersson:

This introduces display clock controllers are introduces for SM6115 and
SM8450, and SC8280XP gains a GPU clock controller.  MSM8909 and SM6375
gains global and SMD RPM clock controller drivers.

The handling of GDSCs with PWRSTS_RET was fixed, to keep the GDSC on
while powering down the parent supply. This solved retention issues
during suspend of USB on sc7180/7280 and SC8280XP.

SM6115 and QCM2260 are moved to reuse PLL configuration. SDM660 SDCC1
was moved to floor ops.

Support for the APCS PLLs for IPQ8064, IPQ8074 and IPQ6018 was
added/fixed. The MSM8996 CPU clocks was updated, with support for ACD
clocks added.

Support for SDM670 was added to the SDM845 Glbal clock controller and
the RPMh clock controller driver.

Transition to parent_data, parent_hws and use of ARRAY_SIZE() for
num_parents was done for MSM8660, MSM8916, MSM8939, MSM8960 global clock
controllers, IPQ8064 LPASS clock controller and MSM8960 multimedia clock
controller.

Support for per-reset defined delay of was introduced.

* tag 'qcom-clk-for-6.1' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux: (93 commits)
  clk: qcom: gcc-msm8939: use ARRAY_SIZE instead of specifying num_parents
  clk: qcom: gcc-msm8939: use parent_hws where possible
  dt-bindings: clock: move qcom,gcc-msm8939 to qcom,gcc-msm8916.yaml
  clk: qcom: gcc-sm6350: Update the .pwrsts for usb gdscs
  clk: qcom: gcc-sc8280xp: use retention for USB power domains
  clk: qcom: gdsc: add missing error handling
  dt-bindings: clocks: qcom,gcc-sc8280xp: Fix typos
  clk: qcom: Add global clock controller driver for SM6375
  dt-bindings: clock: add SM6375 QCOM global clock bindings
  clk: qcom: alpha: Add support for programming the PLL_FSM_LEGACY_MODE bit
  clk: qcom: gcc-sc7280: Update the .pwrsts for usb gdscs
  clk: qcom: gcc-sc7180: Update the .pwrsts for usb gdsc
  clk: qcom: gdsc: Fix the handling of PWRSTS_RET support
  clk: qcom: Add SC8280XP GPU clock controller
  dt-bindings: clock: Add Qualcomm SC8280XP GPU binding
  clk: qcom: smd: Add SM6375 clocks
  dt-bindings: clock: qcom: rpmcc: Add BIMC_FREQ_LOG
  dt-bindings: clock: qcom,rpmcc: Add compatible for SM6375
  clk: qcom: rpmhcc: add sdm670 clocks
  dt-bindings: clock: add rpmhcc bindings for sdm670
  ...

2 years agoclk: allow building lan966x as a module
Clément Léger [Fri, 17 Jun 2022 10:33:06 +0000 (12:33 +0200)]
clk: allow building lan966x as a module

Set the COMMON_CLK_LAN966X option as a tristate and switch from
builtin_platform_driver() to module_platform_driver() to allow building
and using this driver as a module.

Signed-off-by: Clément Léger <clement.leger@bootlin.com>
Link: https://lore.kernel.org/r/20220617103306.489466-1-clement.leger@bootlin.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: clk-xgene: simplify if-if to if-else
Yihao Han [Fri, 8 Apr 2022 13:06:09 +0000 (06:06 -0700)]
clk: clk-xgene: simplify if-if to if-else

Replace `if (!pclk->param.csr_reg)` with `else` for simplification
and add curly brackets according to the kernel coding style:

"Do not unnecessarily use braces where a single statement will do."

...

"This does not apply if only one branch of a conditional statement is
a single statement; in the latter case use braces in both branches"

Please refer to:
https://www.kernel.org/doc/html/v5.17-rc8/process/coding-style.html

Signed-off-by: Yihao Han <hanyihao@vivo.com>
Link: https://lore.kernel.org/r/20220408130617.14963-1-hanyihao@vivo.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: ast2600: BCLK comes from EPLL
Joel Stanley [Thu, 21 Apr 2022 04:04:26 +0000 (13:34 +0930)]
clk: ast2600: BCLK comes from EPLL

This correction was made in the u-boot SDK recently. There are no
in-tree users of this clock so the impact is minimal.

Fixes: d3d04f6c330a ("clk: Add support for AST2600 SoC")
Link: https://github.com/AspeedTech-BMC/u-boot/commit/8ad54a5ae15f27fea5e894cc2539a20d90019717
Signed-off-by: Joel Stanley <joel@jms.id.au>
Link: https://lore.kernel.org/r/20220421040426.171256-1-joel@jms.id.au
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: clocking-wizard: Depend on HAS_IOMEM
Stephen Boyd [Mon, 3 Oct 2022 20:26:08 +0000 (13:26 -0700)]
clk: clocking-wizard: Depend on HAS_IOMEM

This driver uses devm_platform_ioremap_resource() and thus depends on
HAS_IOMEM. Add the Kconfig dependency to avoid build issues.

Reported-by: kernel test robot <lkp@intel.com>
Cc: Shubhrajyoti Datta <shubhrajyoti.datta@xilinx.com>
Fixes: c822490f52da ("clk: clocking-wizard: Move clocking-wizard out")
Link: https://lore.kernel.org/r/20221003202608.2611295-1-sboyd@kernel.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: clocking-wizard: Use dev_err_probe() helper
Yang Yingliang [Tue, 13 Sep 2022 03:14:42 +0000 (11:14 +0800)]
clk: clocking-wizard: Use dev_err_probe() helper

dev_err() can be replace with dev_err_probe() which will check if error
code is -EPROBE_DEFER.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20220913031442.980720-1-yangyingliang@huawei.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: nxp: fix typo in comment
Julia Lawall [Sat, 21 May 2022 11:10:52 +0000 (13:10 +0200)]
clk: nxp: fix typo in comment

Spelling mistake (triple letters) in comment.
Detected with the help of Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Link: https://lore.kernel.org/r/20220521111145.81697-42-Julia.Lawall@inria.fr
Acked-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: pxa: add a check for the return value of kzalloc()
Xiaoke Wang [Thu, 7 Apr 2022 09:33:49 +0000 (17:33 +0800)]
clk: pxa: add a check for the return value of kzalloc()

kzalloc() is a memory allocation function which can return NULL when
some internal memory errors happen. So it is better to check it to
prevent potential wrong memory access.

Signed-off-by: Xiaoke Wang <xkernel.wang@foxmail.com>
Link: https://lore.kernel.org/r/tencent_2B9817738F38B02844C245946EFF3B407E09@qq.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: vc5: Add support for IDT/Renesas VersaClock 5P49V6975
Matthias Fend [Wed, 11 May 2022 05:34:55 +0000 (07:34 +0200)]
clk: vc5: Add support for IDT/Renesas VersaClock 5P49V6975

Update IDT VersaClock 5 driver to support 5P49V6975. The 5P49V6975 is a
member of the VersaClock 6E family and supports four fractional dividers
(FODs), five clock outputs and an internal oscillator.

Signed-off-by: Matthias Fend <matthias.fend@emfend.at>
Link: https://lore.kernel.org/r/20220511053455.360335-2-matthias.fend@emfend.at
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agodt-bindings: clock: vc5: Add 5P49V6975
Matthias Fend [Wed, 11 May 2022 05:34:54 +0000 (07:34 +0200)]
dt-bindings: clock: vc5: Add 5P49V6975

The 5P49V6975 is a member of the VersaClock 6E family and supports four
fractional dividers (FODs), five clock outputs and an internal oscillator.

Signed-off-by: Matthias Fend <matthias.fend@emfend.at>
Link: https://lore.kernel.org/r/20220511053455.360335-1-matthias.fend@emfend.at
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mvebu: armada-37xx-tbg: Remove the unneeded result variable
ye xingchen [Tue, 6 Sep 2022 07:23:22 +0000 (07:23 +0000)]
clk: mvebu: armada-37xx-tbg: Remove the unneeded result variable

Return the value of_clk_add_hw_provider() directly instead of storing it
in another redundant variable.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: ye xingchen <ye.xingchen@zte.com.cn>
Link: https://lore.kernel.org/r/20220906072322.337253-1-ye.xingchen@zte.com.cn
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: ti: dra7-atl: Fix reference leak in of_dra7_atl_clk_probe
Miaoqian Lin [Thu, 2 Jun 2022 03:08:36 +0000 (07:08 +0400)]
clk: ti: dra7-atl: Fix reference leak in of_dra7_atl_clk_probe

pm_runtime_get_sync() will increment pm usage counter.
Forgetting to putting operation will result in reference leak.
Add missing pm_runtime_put_sync in some error paths.

Fixes: 9ac33b0ce81f ("CLK: TI: Driver for DRA7 ATL (Audio Tracking Logic)")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Link: https://lore.kernel.org/r/20220602030838.52057-1-linmq006@gmail.com
Reviewed-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: Renesas versaclock7 ccf device driver
Alex Helms [Mon, 12 Sep 2022 18:36:13 +0000 (11:36 -0700)]
clk: Renesas versaclock7 ccf device driver

Renesas Versaclock7 is a family of configurable clock generator ICs
with fractional and integer dividers. This driver has basic support
for the RC21008A device, a clock synthesizer with a crystal input and
8 outputs. The supports changing the FOD and IOD rates, and each
output can be gated.

Signed-off-by: Alex Helms <alexander.helms.jy@renesas.com>
Link: https://lore.kernel.org/r/20220912183613.22213-3-alexander.helms.jy@renesas.com
Tested-by: Saeed Nowshadi <saeed.nowshadi@amd.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agodt-bindings: Renesas versaclock7 device tree bindings
Alex Helms [Mon, 12 Sep 2022 18:36:12 +0000 (11:36 -0700)]
dt-bindings: Renesas versaclock7 device tree bindings

Renesas Versaclock7 is a family of configurable clock generator ICs
with fractional and integer dividers. This driver has basic support
for the RC21008A device, a clock synthesizer with a crystal input and
8 outputs. The supports changing the FOD and IOD rates, and each
output can be gated.

Signed-off-by: Alex Helms <alexander.helms.jy@renesas.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20220912183613.22213-2-alexander.helms.jy@renesas.com
Tested-by: Saeed Nowshadi <saeed.nowshadi@amd.com>
[sboyd@kernel.org: Rename nodes in example to generic names]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: ti: Balance of_node_get() calls for of_find_node_by_name()
Liang He [Thu, 15 Sep 2022 03:11:21 +0000 (11:11 +0800)]
clk: ti: Balance of_node_get() calls for of_find_node_by_name()

In ti_find_clock_provider(), of_find_node_by_name() will call
of_node_put() for the 'from' argument, possibly putting the node one too
many times. Let's maintain the of_node_get() from the previous search
and only put when we're exiting the function early. This should avoid a
misbalanced reference count on the node.

Fixes: 51f661ef9a10 ("clk: ti: Add ti_find_clock_provider() to use clock-output-names")
Signed-off-by: Liang He <windhl@126.com>
Link: https://lore.kernel.org/r/20220915031121.4003589-1-windhl@126.com
[sboyd@kernel.org: Rewrite commit text, maintain reference instead of
get again]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: imx: scu: fix memleak on platform_device_add() fails
Lin Yujun [Wed, 14 Sep 2022 03:32:06 +0000 (11:32 +0800)]
clk: imx: scu: fix memleak on platform_device_add() fails

No error handling is performed when platform_device_add()
fails. Add error processing before return, and modified
the return value.

Fixes: 77d8f3068c63 ("clk: imx: scu: add two cells binding support")
Signed-off-by: Lin Yujun <linyujun809@huawei.com>
Link: https://lore.kernel.org/r/20220914033206.98046-1-linyujun809@huawei.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: vc5: Use regmap_{set,clear}_bits() where appropriate
Lars-Peter Clausen [Tue, 19 Jul 2022 09:46:37 +0000 (11:46 +0200)]
clk: vc5: Use regmap_{set,clear}_bits() where appropriate

regmap_set_bits() and regmap_clear_bits() are variations of
regmap_update_bits() that can be used if all bits of the mask have to be
set to either 1 or 0 respectively.

Update the versaclk driver to use regmap_set_bits() and regmap_clear_bits()
where appropriate. This results in slightly more compact code and also
makes the intention of the code clearer which can help with review.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20220719094637.844946-2-lars@metafoo.de
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: vc5: Check IO access results
Lars-Peter Clausen [Tue, 19 Jul 2022 09:46:36 +0000 (11:46 +0200)]
clk: vc5: Check IO access results

The devices of the versaclk clock generator family use an I2C control bus.
IO access on an I2C bus can fail for various reasons.

The driver currently ignores the return value of most IO operations. This
results in silent failure. To avoid this check the return value and in case
of an error abort the operation and propagate the error code to the caller.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20220719094637.844946-1-lars@metafoo.de
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mediatek: add driver for MT8365 SoC
Fabien Parent [Mon, 22 Aug 2022 15:26:52 +0000 (17:26 +0200)]
clk: mediatek: add driver for MT8365 SoC

Add clock drivers for MT8365 SoC.

Signed-off-by: Fabien Parent <fparent@baylibre.com>
Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Link: https://lore.kernel.org/r/20220822152652.3499972-5-msp@baylibre.com
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mediatek: Export required common code symbols
Markus Schneider-Pargmann [Mon, 22 Aug 2022 15:26:51 +0000 (17:26 +0200)]
clk: mediatek: Export required common code symbols

To make clk-mt8365 compilable as a module there are a few function
symbols missing. This patch adds the required EXPORT_SYMBOL_GPL to the
functions.

Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Link: https://lore.kernel.org/r/20220822152652.3499972-4-msp@baylibre.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mediatek: Provide mtk_devm_alloc_clk_data
Markus Schneider-Pargmann [Mon, 22 Aug 2022 15:26:50 +0000 (17:26 +0200)]
clk: mediatek: Provide mtk_devm_alloc_clk_data

Provide a helper that replaces the kzalloc with devm_kzalloc so error
handling gets easier.

Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Link: https://lore.kernel.org/r/20220822152652.3499972-3-msp@baylibre.com
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agodt-bindings: clock: mediatek: add bindings for MT8365 SoC
Fabien Parent [Mon, 22 Aug 2022 15:26:49 +0000 (17:26 +0200)]
dt-bindings: clock: mediatek: add bindings for MT8365 SoC

Add the clock bindings for the MediaTek MT8365 SoC.

Signed-off-by: Fabien Parent <fparent@baylibre.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Link: https://lore.kernel.org/r/20220822152652.3499972-2-msp@baylibre.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclkdev: Simplify devm_clk_hw_register_clkdev() function
Andy Shevchenko [Thu, 23 Jun 2022 11:57:19 +0000 (14:57 +0300)]
clkdev: Simplify devm_clk_hw_register_clkdev() function

Use devm_add_action_or_reset() instead of devres_alloc() and
devres_add(), which works the same. This will simplify the
code. There is no functional changes.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20220623115719.52683-3-andriy.shevchenko@linux.intel.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclkdev: Remove never used devm_clk_release_clkdev()
Andy Shevchenko [Thu, 23 Jun 2022 11:57:18 +0000 (14:57 +0300)]
clkdev: Remove never used devm_clk_release_clkdev()

For the entire history of the devm_clk_release_clkdev() existence
(since 2018) it was never used. Remove it for good.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20220623115719.52683-2-andriy.shevchenko@linux.intel.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: Remove never used devm_of_clk_del_provider()
Andy Shevchenko [Thu, 23 Jun 2022 11:57:17 +0000 (14:57 +0300)]
clk: Remove never used devm_of_clk_del_provider()

For the entire history of the devm_of_clk_del_provider) existence
(since 2017) it was never used. Remove it for good.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20220623115719.52683-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: bcm: rpi: Add support for VEC clock
Dom Cobley [Mon, 29 Aug 2022 15:21:54 +0000 (18:21 +0300)]
clk: bcm: rpi: Add support for VEC clock

Platform driver clk-bcm2835 gets an inaccurate clock for VEC (107MHz).
Export VEC clock trough clk-raspberrypi which uses the right PLL to
get an accurate 108MHz.

Signed-off-by: Dom Cobley <popcornmix@gmail.com>
[iivanov: Adapted on top of v5.17-rc6]
Signed-off-by: Ivan T. Ivanov <iivanov@suse.de>
Link: https://lore.kernel.org/r/20220829152154.147250-4-iivanov@suse.de
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: bcm: rpi: Handle pixel clock in firmware
Ivan T. Ivanov [Mon, 29 Aug 2022 15:21:53 +0000 (18:21 +0300)]
clk: bcm: rpi: Handle pixel clock in firmware

The clk-bcm2835 handling of the pixel clock does not function
correctly when the HDMI power domain is disabled.

The firmware supports it correctly, so add it to the
firmware clock driver.

Acked-by: Maxime Ripard <maxime@cerno.tech>
Acked-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Signed-off-by: Ivan T. Ivanov <iivanov@suse.de>
Link: https://lore.kernel.org/r/20220829152154.147250-3-iivanov@suse.de
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: bcm: rpi: Add support HEVC clock
Ivan T. Ivanov [Mon, 29 Aug 2022 15:21:52 +0000 (18:21 +0300)]
clk: bcm: rpi: Add support HEVC clock

Export clock required for RPiVid video decoder hardware.

Cc: Dom Cobley <popcornmix@gmail.com>
Acked-by: Maxime Ripard <maxime@cerno.tech>
Signed-off-by: Ivan T. Ivanov <iivanov@suse.de>
Link: https://lore.kernel.org/r/20220829152154.147250-2-iivanov@suse.de
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: bcm2835: fix bcm2835_clock_rate_from_divisor declaration
Stefan Wahren [Sun, 4 Sep 2022 14:10:37 +0000 (16:10 +0200)]
clk: bcm2835: fix bcm2835_clock_rate_from_divisor declaration

The return value of bcm2835_clock_rate_from_divisor is always unsigned
and also all caller expect this. So fix the declaration accordingly.

Fixes: 41691b8862e2 ("clk: bcm2835: Add support for programming the audio domain clocks")
Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com>
Link: https://lore.kernel.org/r/20220904141037.38816-1-stefan.wahren@i2se.com
Reviewed-by: Ivan T. Ivanov <iivanov@suse.de>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: bcm2835: Round UART input clock up
Ivan T. Ivanov [Mon, 12 Sep 2022 08:13:04 +0000 (11:13 +0300)]
clk: bcm2835: Round UART input clock up

It was reported that RPi3[1] and RPi Zero 2W boards have issues with
the Bluetooth. It turns out that when switching from initial to
operation speed host and device no longer can talk each other because
host uses incorrect UART baud rate.

The UART driver used in this case is amba-pl011. Original fix, see
below Github link[2], was inside pl011 module, but somehow it didn't
look as the right place to fix. Beside that this original rounding
function is not exactly perfect for all possible clock values. So I
deiced to move the hack to the platform which actually need it.

The UART clock is initialised to be as close to the requested
frequency as possible without exceeding it. Now that there is a
clock manager that returns the actual frequencies, an expected
48MHz clock is reported as 47999625. If the requested baud rate
== requested clock/16, there is no headroom and the slight
reduction in actual clock rate results in failure.

If increasing a clock by less than 0.1% changes it from ..999..
to ..000.., round it up.

[1] https://bugzilla.suse.com/show_bug.cgi?id=1188238
[2] https://github.com/raspberrypi/linux/commit/ab3f1b39537f6d3825b8873006fbe2fc5ff057b7

Cc: Phil Elwell <phil@raspberrypi.com>
Signed-off-by: Ivan T. Ivanov <iivanov@suse.de>
Reviewed-by: Stefan Wahren <stefan.wahren@i2se.com>
Link: https://lore.kernel.org/r/20220912081306.24662-1-iivanov@suse.de
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: bcm2835: Make peripheral PLLC critical
Maxime Ripard [Mon, 26 Sep 2022 08:45:09 +0000 (10:45 +0200)]
clk: bcm2835: Make peripheral PLLC critical

When testing for a series affecting the VEC, it was discovered that
turning off and on the VEC clock is crashing the system.

It turns out that, when disabling the VEC clock, it's the only child of
the PLLC-per clock which will also get disabled. The source of the crash
is PLLC-per being disabled.

It's likely that some other device might not take a clock reference that
it actually needs, but it's unclear which at this point. Let's make
PLLC-per critical so that we don't have that crash.

Reported-by: Noralf Trønnes <noralf@tronnes.org>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://lore.kernel.org/r/20220926084509.12233-1-maxime@cerno.tech
Reviewed-by: Stefan Wahren <stefan.wahren@i2se.com>
Acked-by: Noralf Trønnes <noralf@tronnes.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: baikal-t1: Convert to platform device driver
Serge Semin [Thu, 29 Sep 2022 22:54:02 +0000 (01:54 +0300)]
clk: baikal-t1: Convert to platform device driver

In accordance with the way the MIPS platform is normally design there are
only six clock sources which need to be available on the kernel start in
order to one end up booting correctly:
+ CPU PLL: needed by the r4k and MIPS GIC timer drivers. The former one is
  initialized by the arch code, while the later one is implemented in the
  mips-gic-timer.c driver as the OF-declared timer.
+ PCIe PLL: required as a parental clock source for the APB/timer domains.
+ APB clock: needed in order to access all the SoC CSRs at least for the
  timer OF-declared drivers.
+ APB Timer{0-2} clocks: these are the DW APB timers which drivers
  dw_apb_timer_of.c are implemented as the OF-declared timers.

So as long as the clocks above are available early the kernel will
normally work. Let's convert the Baikal-T1 CCU drivers to the platform
device drivers keeping that in mind.

Generally speaking the conversion isn't that complicated since the driver
infrastructure has been designed as flexible enough for that. First we
need to add a new PLL/Divider clock features flag which indicates the
corresponding clock source as a basic one and that clock sources will be
available on the kernel early boot stages. Second the internal PLL/Divider
descriptors need to be initialized with -EPROBE_DEFER value as the
corresponding clock source is unavailable at the early stages. They will
be allocated and initialized on the Baikal-T1 clock platform driver probe
procedure. Finally the already available PLL/Divider init functions need
to be split up into two ones: init procedure performed in the framework of
the OF-declared clock initialization (of_clk_init()), and the probe
procedure called by the platform devices bus driver. Note the later method
will just continue the system clocks initialization started in the former
one.

Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Link: https://lore.kernel.org/r/20220929225402.9696-9-Sergey.Semin@baikalelectronics.ru
[sboyd@kernel.org: Remove module things because the Kconfig is still
bool]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: baikal-t1: Add DDR/PCIe directly controlled resets support
Serge Semin [Thu, 29 Sep 2022 22:54:01 +0000 (01:54 +0300)]
clk: baikal-t1: Add DDR/PCIe directly controlled resets support

Aside with a set of the trigger-like resets Baikal-T1 CCU provides two
additional blocks with directly controlled reset signals. In particular it
concerns DDR full and initial resets and various PCIe sub-domains resets.
Let's add the direct reset assertion/de-assertion of the corresponding
flags support into the Baikal-T1 CCU driver then. It will be required at
least for the PCIe platform driver. Obviously the DDR controller isn't
supposed to be fully reset in the kernel, so the corresponding controls
are added just for the sake of the interface implementation completeness.

Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Link: https://lore.kernel.org/r/20220929225402.9696-8-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agodt-bindings: clk: baikal-t1: Add DDR/PCIe reset IDs
Serge Semin [Thu, 29 Sep 2022 22:54:00 +0000 (01:54 +0300)]
dt-bindings: clk: baikal-t1: Add DDR/PCIe reset IDs

Aside with a set of the trigger-like resets Baikal-T1 CCU provides
additional directly controlled reset signals for the DDR and PCIe
controllers. As a preparation before adding these resets support to the
kernel let's extent the Baikal-T1 CCU IDs list with the new IDs, which
will be used to access the corresponding reset controls.

Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20220929225402.9696-7-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: baikal-t1: Move reset-controls code into a dedicated module
Serge Semin [Thu, 29 Sep 2022 22:53:59 +0000 (01:53 +0300)]
clk: baikal-t1: Move reset-controls code into a dedicated module

Before adding the directly controlled resets support it's reasonable to
move the existing resets control functionality into a dedicated object for
the sake of the CCU dividers clock driver simplification. After the new
functionality was added clk-ccu-div.c would have got to a mixture of the
weakly dependent clocks and resets methods. Splitting the methods up into
the two objects will make the code easier to read and maintain. It shall
also improve the code scalability (though hopefully we won't need this
part that much in the future).

The reset control functionality is now implemented in the framework of a
single unit since splitting it up doesn't make much sense due to
relatively simple reset operations. The ccu-rst.c has been designed to be
looking like ccu-div.c or ccu-pll.c with two globally available methods
for the sake of the code unification and better code readability.

This commit doesn't provide any change in the CCU reset implementation
semantics. As before the driver will support the trigger-like CCU resets
only, which are responsible for the AXI-bus, APB-bus and SATA-ref blocks
reset. The assert/de-assert-capable reset controls support will be added
in the next commit.

Note the CCU Clock dividers and resets functionality split up was possible
due to not having any side-effects (at least we didn't found ones) of the
regmap-based concurrent access of the common CCU dividers/reset CSRs.

Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Link: https://lore.kernel.org/r/20220929225402.9696-6-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: baikal-t1: Add SATA internal ref clock buffer
Serge Semin [Thu, 29 Sep 2022 22:53:58 +0000 (01:53 +0300)]
clk: baikal-t1: Add SATA internal ref clock buffer

It turns out the internal SATA reference clock signal will stay
unavailable for the SATA interface consumer until the buffer on it's way
is ungated. So aside with having the actual clock divider enabled we need
to ungate a buffer placed on the signal way to the SATA controller (most
likely some rudiment from the initial SoC release). Seeing the switch flag
is placed in the same register as the SATA-ref clock divider at a
non-standard ffset, let's implement it as a separate clock controller with
the set-rate propagation to the parental clock divider wrapper. As such
we'll be able to disable/enable and still change the original clock source
rate.

Fixes: 353afa3a8d2e ("clk: Add Baikal-T1 CCU Dividers driver")
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Link: https://lore.kernel.org/r/20220929225402.9696-5-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: baikal-t1: Add shared xGMAC ref/ptp clocks internal parent
Serge Semin [Thu, 29 Sep 2022 22:53:57 +0000 (01:53 +0300)]
clk: baikal-t1: Add shared xGMAC ref/ptp clocks internal parent

Baikal-T1 CCU reference manual says that both xGMAC reference and xGMAC
PTP clocks are generated by two different wrappers with the same constant
divider thus each producing a 156.25 MHz signal. But for some reason both
of these clock sources are gated by a single switch-flag in the CCU
registers space - CCU_SYS_XGMAC_BASE.BIT(0). In order to make the clocks
handled independently we need to define a shared parental gate so the base
clock signal would be switched off only if both of the child-clocks are
disabled.

Note the ID is intentionally set to -2 since we are going to add a one
more internal clock identifier in the next commit.

Fixes: 353afa3a8d2e ("clk: Add Baikal-T1 CCU Dividers driver")
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Link: https://lore.kernel.org/r/20220929225402.9696-4-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: baikal-t1: Fix invalid xGMAC PTP clock divider
Serge Semin [Thu, 29 Sep 2022 22:53:56 +0000 (01:53 +0300)]
clk: baikal-t1: Fix invalid xGMAC PTP clock divider

Most likely due to copy-paste mistake the divider has been set to 10 while
according to the SoC reference manual it's supposed to be 8 thus having
PTP clock frequency of 156.25 MHz.

Fixes: 353afa3a8d2e ("clk: Add Baikal-T1 CCU Dividers driver")
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Link: https://lore.kernel.org/r/20220929225402.9696-3-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: vc5: Fix 5P49V6901 outputs disabling when enabling FOD
Serge Semin [Thu, 29 Sep 2022 22:53:55 +0000 (01:53 +0300)]
clk: vc5: Fix 5P49V6901 outputs disabling when enabling FOD

We have discovered random glitches during the system boot up procedure.
The problem investigation led us to the weird outcomes: when none of the
Renesas 5P49V6901 ports are explicitly enabled by the kernel driver, the
glitches disappeared. It was a mystery since the SoC external clock
domains were fed with different 5P49V6901 outputs. The driver code didn't
seem like bogus either. We almost despaired to find out a root cause when
the solution has been found for a more modern revision of the chip. It
turned out the 5P49V6901 clock generator stopped its output for a short
period of time during the VC5_OUT_DIV_CONTROL register writing. The same
problem was found for the 5P49V6965 revision of the chip and was
successfully fixed in commit fc336ae622df ("clk: vc5: fix output disabling
when enabling a FOD") by enabling the "bypass_sync" flag hidden inside
"Unused Factory Reserved Register". Even though the 5P49V6901 registers
description and programming guide doesn't provide any intel regarding that
flag, setting it up anyway in the officially unused register completely
eliminated the denoted glitches. Thus let's activate the functionality
submitted in commit fc336ae622df ("clk: vc5: fix output disabling when
enabling a FOD") for the Renesas 5P49V6901 chip too in order to remove the
ports implicit inter-dependency.

Fixes: dbf6b16f5683 ("clk: vc5: Add support for IDT VersaClock 5P49V6901")
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Link: https://lore.kernel.org/r/20220929225402.9696-2-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: davinci: cfgchip: Use dev_err_probe() helper
Yang Yingliang [Tue, 13 Sep 2022 03:22:28 +0000 (11:22 +0800)]
clk: davinci: cfgchip: Use dev_err_probe() helper

dev_err() can be replace with dev_err_probe() which will check if error
code is -EPROBE_DEFER.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20220913032228.985852-1-yangyingliang@huawei.com
Reviewed-by: David Lechner <david@lechnology.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: davinci: pll: fix spelling typo in comment
Jiangshan Yi [Mon, 5 Sep 2022 06:58:33 +0000 (14:58 +0800)]
clk: davinci: pll: fix spelling typo in comment

Fix spelling typo in comment.

Reported-by: k2ci <kernel-bot@kylinos.cn>
Signed-off-by: Jiangshan Yi <yijiangshan@kylinos.cn>
Link: https://lore.kernel.org/r/20220905065833.1831473-1-13667453960@163.com
Reviewed-by: David Lechner <david@lechnology.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoMAINTAINERS: add header file to TI DAVINCI SERIES CLOCK DRIVER
Lukas Bulwahn [Wed, 20 Jul 2022 11:00:26 +0000 (13:00 +0200)]
MAINTAINERS: add header file to TI DAVINCI SERIES CLOCK DRIVER

While creating a patch submission on the davinci clock drivers, I noticed
that the header file include/linux/clk/davinci.h belongs to the section
TI DAVINCI SERIES CLOCK DRIVER.

Add a file entry for this header file in TI DAVINCI SERIES CLOCK DRIVER.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Link: https://lore.kernel.org/r/20220720110026.9173-1-lukas.bulwahn@gmail.com
Reviewed-by: David Lechner <david@lechnology.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: control shared SDH bits with separate clock
Doug Brown [Sun, 12 Jun 2022 19:29:37 +0000 (12:29 -0700)]
clk: mmp: pxa168: control shared SDH bits with separate clock

The PXA168 has a peculiar setup with the AXI clock enable control for
the SDHC controllers. The bits in the SDH0 register control the AXI
clock enable for both SDH0 and SDH1. Likewise, the bits in the SDH2
register control both SDH2 and SDH3. This is modeled with two new
parentless clocks that control the shared bits.

Previously, SDH0 had to be enabled in order for SDH1 to be used, and
when SDH1 was enabled, unused bits in the SDH1 register were being
controlled. This fixes those issues. A future commit will add support
for these new shared clocks to be enabled by the PXA168 SDHC driver.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-13-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agodt-bindings: marvell,pxa168: add clock ids for SDH AXI clocks
Doug Brown [Sun, 12 Jun 2022 19:29:36 +0000 (12:29 -0700)]
dt-bindings: marvell,pxa168: add clock ids for SDH AXI clocks

These are clocks shared by SDH0/1 and SDH2/3, respectively.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-12-doug@schmorgal.com
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: add clocks for SDH2 and SDH3
Doug Brown [Sun, 12 Jun 2022 19:29:35 +0000 (12:29 -0700)]
clk: mmp: pxa168: add clocks for SDH2 and SDH3

The PXA168 has four SDHC peripherals. This commit adds the last two.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-11-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agodt-bindings: marvell,pxa168: add clock id for SDH3
Doug Brown [Sun, 12 Jun 2022 19:29:34 +0000 (12:29 -0700)]
dt-bindings: marvell,pxa168: add clock id for SDH3

There are four SDHC peripherals on the PXA168, but only three of them
were present in the DT bindings. This commit adds the fourth.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-10-doug@schmorgal.com
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: fix GPIO clock enable bits
Doug Brown [Sun, 12 Jun 2022 19:29:33 +0000 (12:29 -0700)]
clk: mmp: pxa168: fix GPIO clock enable bits

According to the datasheet, only bit 0 of APBC_GPIO should be controlled
for the clock enable. Bit 1 is marked as reserved (always write 0).

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-9-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: add muxes for more peripherals
Doug Brown [Sun, 12 Jun 2022 19:29:32 +0000 (12:29 -0700)]
clk: mmp: pxa168: add muxes for more peripherals

The TWSI, KPC, PWM, and DFC peripherals didn't have their muxes modeled
in the code, but the PXA168 datasheet shows that they are indeed muxed:

- TWSI can be 31.2 MHz or 62.4 MHz
- KPC can be 32 kHz, 16 kHz, or 26 MHz
- PWM can be 13 MHz or 32 kHz
- DFC can be 156 MHz or 78 MHz

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-8-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: fix incorrect parent clocks
Doug Brown [Sun, 12 Jun 2022 19:29:31 +0000 (12:29 -0700)]
clk: mmp: pxa168: fix incorrect parent clocks

The UART, SDHC, LCD, and CCIC peripherals' muxed parent clocks didn't
match the information provided by the PXA168 datasheet:

- The UART clocks can be 58.5 MHz or the UART PLL. Previously, the first
  mux option was being calculated as 117 MHz, confirmed on hardware to
  be incorrect.

- The SDHC clocks can be 48 MHz, 52 MHz, or 78 MHz. Previously, 48 MHz
  and 52 MHz were swapped. 78 MHz wasn't listed as an option.

- The LCD clock can be 624 MHz or 312 Mhz. Previously, it was being
  calculated as 312 MHz or 52 MHz.

- The CCIC clock can be 156 MHz or 78 MHz. Previously, it was being
  calculated as 312 MHz or 52 MHz.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-7-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: fix const-correctness
Doug Brown [Sun, 12 Jun 2022 19:29:30 +0000 (12:29 -0700)]
clk: mmp: pxa168: fix const-correctness

While working on this series of patches, checkpatch recommended that
an extra const should be added to the mux parent arrays.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-6-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: add new clocks for peripherals
Doug Brown [Sun, 12 Jun 2022 19:29:29 +0000 (12:29 -0700)]
clk: mmp: pxa168: add new clocks for peripherals

This commit adds three new clocks that previously didn't exist, but are
needed in order to match the clock parenting as described in the PXA168
datasheet.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-5-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agodt-bindings: marvell,pxa168: add clock ids for additional dividers
Doug Brown [Sun, 12 Jun 2022 19:29:28 +0000 (12:29 -0700)]
dt-bindings: marvell,pxa168: add clock ids for additional dividers

This adds a few new clocks divided from PLL1 and CLK32 that are
potentially used by a few peripherals with muxed clocks.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-4-doug@schmorgal.com
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: fix incorrect dividers
Doug Brown [Sun, 12 Jun 2022 19:29:27 +0000 (12:29 -0700)]
clk: mmp: pxa168: fix incorrect dividers

These two clocks had multipliers and dividers that didn't match their
names. A subsequent commit goes through all of the existing peripherals
and ensure the correct clocks are being used everywhere.

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-3-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: mmp: pxa168: add additional register defines
Doug Brown [Sun, 12 Jun 2022 19:29:26 +0000 (12:29 -0700)]
clk: mmp: pxa168: add additional register defines

In preparation for adding additional peripherals over time, this commit
adds a bunch of extra APBC_* defines based on information from the
datasheet. It also reorganizes the list of defines to be ordered
sequentially by address (grouped by type).

Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220612192937.162952-2-doug@schmorgal.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: sprd: Add clocks support for UMS512
Cixi Geng [Fri, 9 Sep 2022 15:24:21 +0000 (23:24 +0800)]
clk: sprd: Add clocks support for UMS512

Add the list of clocks for the Unisoc UMS512, along with clock
initialization.

Signed-off-by: Cixi Geng <cixi.geng1@unisoc.com>
Link: https://lore.kernel.org/r/20220909152421.278662-3-gengcixi@gmail.com
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: fixed-rate: add devm_clk_hw_register_fixed_rate
Dmitry Baryshkov [Fri, 16 Sep 2022 06:17:39 +0000 (09:17 +0300)]
clk: fixed-rate: add devm_clk_hw_register_fixed_rate

Add devm_clk_hw_register_fixed_rate(), devres-managed helper to register
fixed-rate clock.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20220916061740.87167-3-dmitry.baryshkov@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: asm9260: use parent index to link the reference clock
Dmitry Baryshkov [Fri, 16 Sep 2022 06:17:38 +0000 (09:17 +0300)]
clk: asm9260: use parent index to link the reference clock

Rewrite clk-asm9260 to use parent index to use the reference clock.
During this rework two helpers are added:

- clk_hw_register_mux_table_parent_data() to supplement
  clk_hw_register_mux_table() but using parent_data instead of
  parent_names

- clk_hw_register_fixed_rate_parent_accuracy() to be used instead of
  directly calling __clk_hw_register_fixed_rate(). The later function is
  an internal API, which is better not to be called directly.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20220916061740.87167-2-dmitry.baryshkov@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoMerge tag 'mtk-clk-for-6.1' of https://git.kernel.org/pub/scm/linux/kernel/git/wens...
Stephen Boyd [Fri, 30 Sep 2022 00:39:10 +0000 (17:39 -0700)]
Merge tag 'mtk-clk-for-6.1' of https://git./linux/kernel/git/wens/linux into clk-mtk

Pull MediaTek clk driver updates from Chen-Yu Tsai:

A lot of clean up work, as well as new drivers and new functions

 - New clock drivers for MediaTek Helio X10 MT6795
 - Add missing DPI1_HDMI clock in MT8195 VDOSYS1
 - Clock driver changes to support GPU DVFS on MT8183, MT8192, MT8195
   - Fix GPU clock topology on MT8195
   - Propogate rate changes from GPU clock gate up the tree
   - Clock mux notifiers for GPU-related PLLs
 - Conversion of more "simple" drivers to mtk_clk_simple_probe()
 - Hook up mtk_clk_simple_remove() for "simple" MT8192 clock drivers
 - Fixes to previous |struct clk| to |struct clk_hw| conversion
 - Shrink MT8192 clock driver by deduplicating clock parent lists

* tag 'mtk-clk-for-6.1' of https://git.kernel.org/pub/scm/linux/kernel/git/wens/linux: (31 commits)
  clk: mediatek: mt8192: deduplicate parent clock lists
  clk: mediatek: Migrate remaining clk_unregister_*() to clk_hw_unregister_*()
  clk: mediatek: fix unregister function in mtk_clk_register_dividers cleanup
  clk: mediatek: clk-mt8192: Add clock mux notifier for mfg_pll_sel
  clk: mediatek: clk-mt8192-mfg: Propagate rate changes to parent
  clk: mediatek: clk-mt8195-topckgen: Drop univplls from mfg mux parents
  clk: mediatek: clk-mt8195-topckgen: Add GPU clock mux notifier
  clk: mediatek: clk-mt8195-topckgen: Register mfg_ck_fast_ref as generic mux
  clk: mediatek: clk-mt8195-mfg: Reparent mfg_bg3d and propagate rate changes
  clk: mediatek: mt8183: Add clk mux notifier for MFG mux
  clk: mediatek: mux: add clk notifier functions
  clk: mediatek: mt8183: mfgcfg: Propagate rate changes to parent
  clk: mediatek: Use mtk_clk_register_gates_with_dev in simple probe
  clk: mediatek: gate: Export mtk_clk_register_gates_with_dev
  clk: mediatek: add VDOSYS1 clock
  dt-bindings: clk: mediatek: Add MT8195 DPI clocks
  clk: mediatek: mt8192: add mtk_clk_simple_remove
  clk: mediatek: mt8183: use mtk_clk_simple_probe to simplify driver
  clk: mediatek: mt6797: use mtk_clk_simple_probe to simplify driver
  clk: mediatek: mt6779: use mtk_clk_simple_probe to simplify driver
  ...

2 years agoclk: qcom: gcc-msm8939: use ARRAY_SIZE instead of specifying num_parents
Dmitry Baryshkov [Wed, 28 Sep 2022 14:56:09 +0000 (17:56 +0300)]
clk: qcom: gcc-msm8939: use ARRAY_SIZE instead of specifying num_parents

Use ARRAY_SIZE() instead of manually specifying num_parents. This makes
adding/removing entries to/from parent_data easy and errorproof.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20220928145609.375860-4-dmitry.baryshkov@linaro.org
2 years agoclk: qcom: gcc-msm8939: use parent_hws where possible
Dmitry Baryshkov [Wed, 28 Sep 2022 14:56:08 +0000 (17:56 +0300)]
clk: qcom: gcc-msm8939: use parent_hws where possible

Use parent_hws instead of hanving parent_data with just a single .hw
entry to speed up and simplify parent lookups.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20220928145609.375860-3-dmitry.baryshkov@linaro.org
2 years agodt-bindings: clock: move qcom,gcc-msm8939 to qcom,gcc-msm8916.yaml
Dmitry Baryshkov [Wed, 28 Sep 2022 14:56:07 +0000 (17:56 +0300)]
dt-bindings: clock: move qcom,gcc-msm8939 to qcom,gcc-msm8916.yaml

The MSM8939 GCC bindings are fully comptible with MSM8916, the clock
controller requires the same parent clocks, move MSM8939 GCC compatible
to qcom,msm8916.yaml

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20220928145609.375860-2-dmitry.baryshkov@linaro.org
2 years agoclk: qcom: gcc-sm6350: Update the .pwrsts for usb gdscs
Luca Weiss [Wed, 28 Sep 2022 13:28:54 +0000 (15:28 +0200)]
clk: qcom: gcc-sm6350: Update the .pwrsts for usb gdscs

The USB controllers on sm6350 do not retain the state when
the system goes into low power state and the GDSCs are
turned off.

This can be observed by the USB connection not coming back alive after
putting the device into suspend, essentially breaking USB.

Fix this by updating the .pwrsts for the USB GDSCs so they only
transition to retention state in low power.

Cc: Rajendra Nayak <quic_rjendra@quicinc.com>
Signed-off-by: Luca Weiss <luca.weiss@fairphone.com>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20220928132853.179425-1-luca.weiss@fairphone.com
2 years agoclk: qcom: gcc-sc8280xp: use retention for USB power domains
Johan Hovold [Thu, 29 Sep 2022 16:11:24 +0000 (18:11 +0200)]
clk: qcom: gcc-sc8280xp: use retention for USB power domains

Since commit d399723950c4 ("clk: qcom: gdsc: Fix the handling of
PWRSTS_RET support) retention mode can be used on sc8280xp to maintain
state during suspend instead of leaving the domain always on.

This is needed to eventually allow the parent CX domain to be powered
down during suspend.

Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20220929161124.18138-1-johan+linaro@kernel.org
2 years agoclk: qcom: gdsc: add missing error handling
Johan Hovold [Thu, 29 Sep 2022 15:58:16 +0000 (17:58 +0200)]
clk: qcom: gdsc: add missing error handling

Since commit 7eb231c337e0 ("PM / Domains: Convert pm_genpd_init() to
return an error code") pm_genpd_init() can return an error which the
caller must handle.

The current error handling was also incomplete as the runtime PM and
regulator use counts were not balanced in all error paths.

Add the missing error handling to the GDSC initialisation to avoid
continuing as if nothing happened on errors.

Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Link: https://lore.kernel.org/r/20220929155816.17425-1-johan+linaro@kernel.org
2 years agoclk: mediatek: mt8192: deduplicate parent clock lists
Chen-Yu Tsai [Mon, 26 Sep 2022 10:25:22 +0000 (18:25 +0800)]
clk: mediatek: mt8192: deduplicate parent clock lists

Some groups of clocks of the same type share the same list of parents.
These lists were declared separately for each clock in older drivers,
bloating the code.

Merge some obvious duplicate parent clock lists in the MT8192 clock
driver together to reduce the code size. These include:

- apll_i2s*_m_parents into one as apll_i2s_m_parents
- img1_parents & img2_parents into one as img_parents
- msdc30_*_parents into one as msdc30_parents
- camtg*_parents into cam_tg_parents
- seninf*_parents into seninf_parents

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20220926102523.2367530-6-wenst@chromium.org
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: Migrate remaining clk_unregister_*() to clk_hw_unregister_*()
Chen-Yu Tsai [Mon, 26 Sep 2022 10:25:19 +0000 (18:25 +0800)]
clk: mediatek: Migrate remaining clk_unregister_*() to clk_hw_unregister_*()

During the previous |struct clk| to |struct clk_hw| clk provider API
migration in commit 6f691a586296 ("clk: mediatek: Switch to clk_hw
provider APIs"), a few clk_unregister_*() calls were missed.

Migrate the remaining ones to the |struct clk_hw| provider API, i.e.
change clk_unregister_*() to clk_hw_unregister_*().

Fixes: 6f691a586296 ("clk: mediatek: Switch to clk_hw provider APIs")
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20220926102523.2367530-3-wenst@chromium.org
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: fix unregister function in mtk_clk_register_dividers cleanup
Chen-Yu Tsai [Mon, 26 Sep 2022 10:25:18 +0000 (18:25 +0800)]
clk: mediatek: fix unregister function in mtk_clk_register_dividers cleanup

When the cleanup paths for the various clk register APIs in the MediaTek
clk library were added, the one in the dividers type used the wrong type
of unregister function. This would result in incorrect dereferencing of
the clk pointer and freeing of invalid pointers.

Fix this by switching to the correct type of clk unregistration call.

Fixes: 3c3ba2ab0226 ("clk: mediatek: mtk: Implement error handling in register APIs")
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20220926102523.2367530-2-wenst@chromium.org
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: clk-mt8192: Add clock mux notifier for mfg_pll_sel
AngeloGioacchino Del Regno [Tue, 27 Sep 2022 10:11:28 +0000 (12:11 +0200)]
clk: mediatek: clk-mt8192: Add clock mux notifier for mfg_pll_sel

Following the changes that were done for mt8183, add a clock notifier
for the GPU PLL selector mux: this allows safe clock rate changes by
temporarily reparenting the GPU to a safe clock (clk26m) while the
MFGPLL is reprogrammed and stabilizes.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20220927101128.44758-11-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: clk-mt8192-mfg: Propagate rate changes to parent
AngeloGioacchino Del Regno [Tue, 27 Sep 2022 10:11:27 +0000 (12:11 +0200)]
clk: mediatek: clk-mt8192-mfg: Propagate rate changes to parent

Following what was done on MT8183 and MT8195, also propagate the rate
changes to MFG_BG3D's parent on MT8192 to allow for proper GPU DVFS.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20220927101128.44758-10-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: clk-mt8195-topckgen: Drop univplls from mfg mux parents
AngeloGioacchino Del Regno [Tue, 27 Sep 2022 10:11:26 +0000 (12:11 +0200)]
clk: mediatek: clk-mt8195-topckgen: Drop univplls from mfg mux parents

These PLLs are conflicting with GPU rates that can be generated by
the GPU-dedicated MFGPLL and would require a special clock handler
to be used, for very little and ignorable power consumption benefits.
Also, we're in any case unable to set the rate of these PLLs to
something else that is sensible for this task, so simply drop them:
this will make the GPU to be clocked exclusively from MFGPLL for
"fast" rates, while still achieving the right "safe" rate during
PLL frequency locking.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20220927101128.44758-9-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: clk-mt8195-topckgen: Add GPU clock mux notifier
AngeloGioacchino Del Regno [Tue, 27 Sep 2022 10:11:25 +0000 (12:11 +0200)]
clk: mediatek: clk-mt8195-topckgen: Add GPU clock mux notifier

Following the changes done to MT8183, register a similar notifier
for MT8195 as well, allowing safe clockrate updates for the MFGPLL.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220927101128.44758-8-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: clk-mt8195-topckgen: Register mfg_ck_fast_ref as generic mux
AngeloGioacchino Del Regno [Tue, 27 Sep 2022 10:11:24 +0000 (12:11 +0200)]
clk: mediatek: clk-mt8195-topckgen: Register mfg_ck_fast_ref as generic mux

This clock was being registered as clk-composite through the helpers
for the same in the MediaTek clock APIs but, in reality, this isn't
a composite clock.

Appropriately register this clock with devm_clk_hw_register_mux().
No functional changes.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20220927101128.44758-7-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: clk-mt8195-mfg: Reparent mfg_bg3d and propagate rate changes
AngeloGioacchino Del Regno [Tue, 27 Sep 2022 10:11:23 +0000 (12:11 +0200)]
clk: mediatek: clk-mt8195-mfg: Reparent mfg_bg3d and propagate rate changes

The MFG_BG3D is a gate to enable/disable clock output to the GPU,
but the actual output is decided by multiple muxes; in particular:
mfg_ck_fast_ref muxes between "slow" (top_mfg_core_tmp) and
"fast" (MFGPLL) clock, while top_mfg_core_tmp muxes between the
26MHz clock and various system PLLs.

The clock gate comes after all the muxes, so its parent is
mfg_ck_fast_reg, not top_mfg_core_tmp.
Reparent MFG_BG3D to the latter to match the hardware and add the
CLK_SET_RATE_PARENT flag to it: this way we ensure propagating
rate changes that are requested on MFG_BG3D along its entire clock
tree.

Fixes: 35016f10c0e5 ("clk: mediatek: Add MT8195 mfgcfg clock support")
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20220927101128.44758-6-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: mt8183: Add clk mux notifier for MFG mux
Chen-Yu Tsai [Tue, 27 Sep 2022 10:11:22 +0000 (12:11 +0200)]
clk: mediatek: mt8183: Add clk mux notifier for MFG mux

When the MFG PLL clock, which is upstream of the MFG clock, is changed,
the downstream clock and consumers need to be switched away from the PLL
over to a stable clock to avoid glitches.

This is done through the use of the newly added clk mux notifier. The
notifier is set on the mux itself instead of the upstream PLL, but in
practice this works, as the rate change notifitcations are propogated
throughout the sub-tree hanging off the PLL. Just before rate changes,
the MFG mux is temporarily and transparently switched to the 26 MHz
main crystal. After the rate change, the mux is switched back.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
[Angelo: Rebased to assign clk_ops in mtk_mux_nb]
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220927101128.44758-5-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: mux: add clk notifier functions
Chen-Yu Tsai [Tue, 27 Sep 2022 10:11:21 +0000 (12:11 +0200)]
clk: mediatek: mux: add clk notifier functions

With device frequency scaling, the mux clock that (indirectly) feeds the
device selects between a dedicated PLL, and some other stable clocks.

When a clk rate change is requested, the (normally) upstream PLL is
reconfigured. It's possible for the clock output of the PLL to become
unstable during this process.

To avoid causing the device to glitch, the mux should temporarily be
switched over to another "stable" clock during the PLL rate change.
This is done with clk notifiers.

This patch adds common functions for notifiers to temporarily and
transparently reparent mux clocks.

This was loosely based on commit 8adfb08605a9 ("clk: sunxi-ng: mux: Add
clk notifier functions").

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
[Angelo: Changed mtk_mux_nb to hold a pointer to clk_ops instead of mtk_mux]
Co-developed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220927101128.44758-4-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: mediatek: mt8183: mfgcfg: Propagate rate changes to parent
Chen-Yu Tsai [Tue, 27 Sep 2022 10:11:20 +0000 (12:11 +0200)]
clk: mediatek: mt8183: mfgcfg: Propagate rate changes to parent

The only clock in the MT8183 MFGCFG block feeds the GPU. Propagate its
rate change requests to its parent, so that DVFS for the GPU can work
properly.

Fixes: acddfc2c261b ("clk: mediatek: Add MT8183 clock support")
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20220927101128.44758-3-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
2 years agoclk: pistachio: Fix initconst confusion
Andi Kleen [Tue, 20 Sep 2022 05:58:38 +0000 (07:58 +0200)]
clk: pistachio: Fix initconst confusion

A variable pointing to const isn't const itself. It has to contain
"const" keyword after "*" too. So to keep it in __initconst (and not
mark properly as __initdata), add the "const" keyword exactly there.

Note we need to update struct pistachio_mux too. On the other hand, the
clk core already counts with "const char *const" already.

[js] more explanatory commit message.

Cc: Michael Turquette <mturquette@baylibre.com>
Cc: Stephen Boyd <sboyd@kernel.org>
Cc: linux-clk@vger.kernel.org
Cc: Martin Liska <mliska@suse.cz>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Link: https://lore.kernel.org/r/20220920055838.22637-1-jslaby@suse.cz
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoclk: clk-npcm7xx: Remove unused struct npcm7xx_clk_gate_data and npcm7xx_clk_div_fixe...
Yuan Can [Tue, 27 Sep 2022 13:39:31 +0000 (13:39 +0000)]
clk: clk-npcm7xx: Remove unused struct npcm7xx_clk_gate_data and npcm7xx_clk_div_fixed_data

After commit 6a5898411159("clk: clk-npcm7xx: Remove unused static const tables
'npcm7xx_gates' and 'npcm7xx_divs_fx'"), no one use struct
npcm7xx_clk_gate_data and struct npcm7xx_clk_div_fixed_data, so remove them.

Signed-off-by: Yuan Can <yuancan@huawei.com>
Link: https://lore.kernel.org/r/20220927133931.104060-1-yuancan@huawei.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2 years agoMerge tag 'clk-imx-6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa...
Stephen Boyd [Wed, 28 Sep 2022 23:25:45 +0000 (16:25 -0700)]
Merge tag 'clk-imx-6.1' of git://git./linux/kernel/git/abelvesa/linux into clk-imx

Pull i.MX clk driver updates from Abel Vesa:

 - Change order between 'sim_enet_root_clk' and 'enet_qos_root_clk'
   clocks for i.MX8MP
 - Drop unnecessary newline in i.MX8MM dt-bindings
 - Add more MU1 and SAI clocks dt-bindings Ids
 - Introduce slice busy bit check for i.MX93 composite clock
 - Introduce white list bit check for i.MX93 composite clock
 - Add new i.MX93 clock gate
 - Add MU1 and MU2 clocks to i.MX93 clock provider
 - Add SAI IPG clocks to i.MX93 clock provider

* tag 'clk-imx-6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa/linux:
  clk: imx93: add SAI IPG clk
  clk: imx93: add MU1/2 clock
  clk: imx93: switch to use new clk gate API
  clk: imx: add i.MX93 clk gate
  clk: imx: clk-composite-93: check white_list
  clk: imx: clk-composite-93: check slice busy
  dt-bindings: clock: imx93-clock: add more MU/SAI clocks
  dt-bindings: clock: imx8mm: don't use multiple blank lines
  clk: imx8mp: tune the order of enet_qos_root_clk

2 years agoMerge tag 'sunxi-clk-for-6.1-1' of https://git.kernel.org/pub/scm/linux/kernel/git...
Stephen Boyd [Wed, 28 Sep 2022 23:21:13 +0000 (16:21 -0700)]
Merge tag 'sunxi-clk-for-6.1-1' of https://git./linux/kernel/git/sunxi/linux into clk-allwinner

Pull Allwinner clk driver updates from Jernej Skrabec:

Refactor A80 DE & USB and DE2 code to use dev_err_probe() helper

* tag 'sunxi-clk-for-6.1-1' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
  clk: sunxi-ng: ccu-sun9i-a80-usb: Use dev_err_probe() helper
  clk: sunxi-ng: ccu-sun9i-a80-de: Use dev_err_probe() helper
  clk: sunxi-ng: sun8i-de2: Use dev_err_probe() helper
  clk: sunxi-ng: d1: Limit PLL rates to stable ranges

2 years agoMerge tag 'clk-microchip-6.1' of https://git.kernel.org/pub/scm/linux/kernel/git...
Stephen Boyd [Wed, 28 Sep 2022 23:10:20 +0000 (16:10 -0700)]
Merge tag 'clk-microchip-6.1' of https://git./linux/kernel/git/at91/linux into clk-microchip

Pull Microchip clk driver updates from Claudiu Beznea:

Microchip AT91:
- add generic clocks for U(S)ART available on SAMA5D2 SoCs

Microchip Polarfire:
- reset controller support for Polarfire clocks
- .round_rate and .set rate support for clk-mpfs
- code cleanup for clk-mpfs
- PLL support for PolarFire SoC's Clock Conditioning Circuitry

* tag 'clk-microchip-6.1' of https://git.kernel.org/pub/scm/linux/kernel/git/at91/linux:
  clk: at91: sama5d2: Add Generic Clocks for UART/USART
  clk: microchip: add PolarFire SoC fabric clock support
  dt-bindings: clk: add PolarFire SoC fabric clock ids
  dt-bindings: clk: document PolarFire SoC fabric clocks
  dt-bindings: clk: rename mpfs-clkcfg binding
  clk: microchip: mpfs: update module authorship & licencing
  clk: microchip: mpfs: convert periph_clk to clk_gate
  clk: microchip: mpfs: convert cfg_clk to clk_divider
  clk: microchip: mpfs: delete 2 line mpfs_clk_register_foo()
  clk: microchip: mpfs: simplify control reg access
  clk: microchip: mpfs: move id & offset out of clock structs
  clk: microchip: mpfs: add MSS pll's set & round rate
  MAINTAINERS: add polarfire soc reset controller
  reset: add polarfire soc reset support
  clk: microchip: mpfs: add reset controller
  dt-bindings: clk: microchip: mpfs: add reset controller support
  clk: microchip: mpfs: make the rtc's ahb clock critical
  clk: microchip: mpfs: fix clk_cfg array bounds violation

2 years agoMerge tag 'renesas-clk-for-v6.1-tag2' of git://git.kernel.org/pub/scm/linux/kernel...
Stephen Boyd [Wed, 28 Sep 2022 23:07:09 +0000 (16:07 -0700)]
Merge tag 'renesas-clk-for-v6.1-tag2' of git://git./linux/kernel/git/geert/renesas-drivers into clk-renesas

Pull some more Renesas clk driver updates from Geert Uytterhoeven:

  - Add watchdog, I2C, pin control/GPIO, and Ethernet clocks on R-Car
    V4H

* tag 'renesas-clk-for-v6.1-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers:
  clk: renesas: r8a779g0: Add EtherAVB clocks
  clk: renesas: r8a779g0: Add PFC/GPIO clocks
  clk: renesas: r8a779g0: Add I2C clocks
  clk: renesas: r8a779g0: Add watchdog clock

2 years agoMerge tag 'renesas-clk-for-v6.1-tag1' of git://git.kernel.org/pub/scm/linux/kernel...
Stephen Boyd [Wed, 28 Sep 2022 23:01:17 +0000 (16:01 -0700)]
Merge tag 'renesas-clk-for-v6.1-tag1' of git://git./linux/kernel/git/geert/renesas-drivers into clk-renesas

Pull Renesas clk driver updates from Geert Uytterhoeven:

  - Add SDHI, Timer (CMT/TMU), and SPI (MSIOF) clocks on R-Car S4-8
  - Add I2C clocks and resets on RZ/V2M
  - Document clock support for the RZ/Five SoC
  - Miscellaneous fixes and improvements

* tag 'renesas-clk-for-v6.1-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers:
  dt-bindings: clock: renesas,rzg2l: Document RZ/Five SoC
  clk: renesas: r8a779f0: Add MSIOF clocks
  clk: renesas: r9a09g011: Add IIC clock and reset entries
  clk: renesas: r9a07g044: Add conditional compilation for r9a07g044_cpg_info
  clk: renesas: r8a779f0: Add TMU and parent SASYNC clocks
  clk: renesas: r8a779f0: Add CMT clocks
  clk: renesas: r8a779f0: Add SDH0 clock

2 years agoMerge tag 'v6.1-rockchip-clock1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Stephen Boyd [Wed, 28 Sep 2022 22:54:35 +0000 (15:54 -0700)]
Merge tag 'v6.1-rockchip-clock1' of git://git./linux/kernel/git/mmind/linux-rockchip into clk-rockchip

Pull Rockchip clk driver updates from Heiko Stuebner:

 - mux-variant clock using the table variant to select parents
 - clock controller for the rv1126 soc
 - conversion of rk3128 to yaml and relicensing of the yaml bindings
   to gpl2+MIT (following dt-binding guildelines)

* tag 'v6.1-rockchip-clock1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:
  dt-bindings: clock: rockchip: change SPDX-License-Identifier
  dt-bindings: clock: convert rockchip,rk3128-cru.txt to YAML
  clk: rockchip: Add clock controller support for RV1126 SoC
  dt-bindings: clock: rockchip: Document RV1126 CRU
  clk: rockchip: Add dt-binding header for RV1126
  clk: rockchip: Add MUXTBL variant