Merge branches 'clk-ux500', 'clk-mtk', 'clk-tegra', 'clk-allwinner' and 'clk-imx' into clk-next

 - Convert ux500 to clk_hw
 - Add the two missing CLKOUT clocks for U8500/DB8500 SoC
 - MediaTek MT8186 SoC clk support
 - Move MediaTek driver to clk_hw provider APIs

* clk-ux500:
  clk: ux500: fix a possible off-by-one in u8500_prcc_reset_base()
  clk: ux500: Implement the missing CLKOUT clocks
  clk: ux500: Rewrite PRCMU clocks to use clk_hw_*
  clk: ux500: Drop .is_prepared state from PRCMU clocks
  clk: ux500: Drop .is_enabled state from PRCMU clocks
  dt-bindings: clock: u8500: Add clkout clock bindings

* clk-mtk: (22 commits)
  clk: mediatek: mt8173: Switch to clk_hw provider APIs
  clk: mediatek: Switch to clk_hw provider APIs
  clk: mediatek: Replace 'struct clk' with 'struct clk_hw'
  clk: mediatek: apmixed: Drop error message from clk_register() failure
  clk: mediatek: Make mtk_clk_register_composite() static
  clk: mediatek: use en_mask as a pure div_en_mask
  clk: mediatek: update compatible string for MT7986 ethsys
  clk: mediatek: Add MT8186 ipesys clock support
  clk: mediatek: Add MT8186 mdpsys clock support
  clk: mediatek: Add MT8186 camsys clock support
  clk: mediatek: Add MT8186 vencsys clock support
  clk: mediatek: Add MT8186 vdecsys clock support
  clk: mediatek: Add MT8186 imgsys clock support
  clk: mediatek: Add MT8186 wpesys clock support
  clk: mediatek: Add MT8186 mmsys clock support
  clk: mediatek: Add MT8186 mfgsys clock support
  clk: mediatek: Add MT8186 imp i2c wrapper clock support
  clk: mediatek: Add MT8186 apmixedsys clock support
  clk: mediatek: Add MT8186 infrastructure clock support
  clk: mediatek: Add MT8186 topckgen clock support
  ...

* clk-tegra:
  clk: tegra: Update kerneldoc to match prototypes
  clk: tegra: Replace .round_rate() with .determine_rate()
  clk: tegra: Register clocks from root to leaf
  clk: tegra: Add missing reset deassertion

* clk-allwinner:
  clk: sunxi-ng: h616: Add PLL derived 32KHz clock
  clk: sunxi-ng: h6-r: Add RTC gate clock

* clk-imx:
  clk: imx8mp: fix usb_root_clk parent
  clk: imx8mp: add clkout1/2 support
  clk: imx: scu: Use pm_runtime_resume_and_get to fix pm_runtime_get_sync() usage
  clk: imx8mp: Add DISP2 pixel clock
  clk: imx: scu: fix a potential memory leak in __imx_clk_gpr_scu()
  clk: imx: Add check for kcalloc
  clk: imx8mn: add GPT support
  dt-bindings: imx: add clock bindings for i.MX8MN GPT
  clk: imx: Remove the snvs clock
  clk: imx8m: check mcore_booted before register clk
  clk: imx: add mcore_booted module paratemter
  clk: imx8mq: add 27m phy pll ref clock

Merge branches 'clk-ti', 'clk-cleanup', 'clk-airoha', 'clk-i2c-simple' and 'clk-renesas' into clk-next

 - Airoha EN7523 SoC system clocks
 - Use i2c driver probe_new to avoid id scans

* clk-ti:
  clk: ti: clkctrl: replace usage of found with dedicated list iterator variable
  clk: ti: composite: Prefer kcalloc over open coded arithmetic
  clk: keystone: syscon-clk: Add support for AM62 epwm-tbclk
  dt-bindings: clock: ehrpwm: Add AM62 specific compatible

* clk-cleanup:
  clk: bcm: rpi: Use correct order for the parameters of devm_kcalloc()
  clk: fixed-rate: Remove redundant if statement
  clk: mux: remove redundant initialization of variable width
  clk: using pm_runtime_resume_and_get instead of pm_runtime_get_sync
  clk: actions: remove redundant assignment after a mask operation

* clk-airoha:
  clk: en7523: fix wrong pointer check in en7523_clk_probe()
  clk: en7523: Add clock driver for Airoha EN7523 SoC
  dt-bindings: Add en7523-scu device tree binding documentation

* clk-i2c-simple:
  clk: renesas-pcie: use simple i2c probe function
  clk: si570: use i2c_match_id and simple i2c probe
  clk: si544: use i2c_match_id and simple i2c probe
  clk: si5351: use i2c_match_id and simple i2c probe
  clk: si5341: use simple i2c probe function
  clk: si514: use simple i2c probe function
  clk: max9485: use simple i2c probe function
  clk: cs2000-cp: use simple i2c probe function
  clk: cdce925: use i2c_match_id and simple i2c probe
  clk: cdce706: use simple i2c probe function

* clk-renesas: (48 commits)
  clk: renesas: r9a09g011: Add eth clock and reset entries
  clk: renesas: Add RZ/V2M support using the rzg2l driver
  clk: renesas: rzg2l: Add support for RZ/V2M reset monitor reg
  clk: renesas: rzg2l: Make use of CLK_MON registers optional
  clk: renesas: rzg2l: Set HIWORD mask for all mux and dividers
  clk: renesas: rzg2l: Add read only versions of the clk macros
  clk: renesas: rzg2l: Move the DEF_MUX array size calc into the macro
  dt-bindings: clock: renesas,rzg2l: Document RZ/V2M SoC
  clk: renesas: r9a07g044: Fix OSTM1 module clock name
  clk: renesas: r9a07g043: Add clock and reset entries for ADC
  clk: renesas: r9a07g043: Add TSU clock and reset entry
  clk: renesas: r9a07g043: Add RSPI clock and reset entries
  clk: renesas: r9a07g043: Add clock and reset entries for SPI Multi I/O Bus Controller
  clk: renesas: r9a07g044: Add DSI clock and reset entries
  clk: renesas: r9a07g044: Add LCDC clock and reset entries
  clk: renesas: r9a07g044: Add M4 Clock support
  clk: renesas: r9a07g044: Add M3 Clock support
  clk: renesas: r9a07g044: Add {M2, M2_DIV2} Clocks support
  clk: renesas: r9a07g044: Add M1 clock support
  clk: renesas: rzg2l: Add DSI divider clk support
  ...

clk: bcm: rpi: Use correct order for the parameters of devm_kcalloc()

We should have 'n', then 'size', not the opposite.
This is harmless because the 2 values are just multiplied, but having
the correct order silence a (unpublished yet) smatch warning.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/r/49d726d11964ca0e3757bdb5659e3b3eaa1572b5.1653081643.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: ti: clkctrl: replace usage of found with dedicated list iterator variable

To move the list iterator variable into the list_for_each_entry_*()
macro in the future it should be avoided to use the list iterator
variable after the loop body.

To *never* use the list iterator variable after the loop it was
concluded to use a separate iterator variable instead of a
found boolean [1].

This removes the need to use a found variable and simply checking if
the variable was set, can determine if the break/goto was hit.

Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/
Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
Link: https://lore.kernel.org/r/20220324071019.59483-1-jakobkoschel@gmail.com
Tested-by: Tony Lindgren <tony@atomide.com>
Reviewed-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: ti: composite: Prefer kcalloc over open coded arithmetic

As noted in the "Deprecated Interfaces, Language Features, Attributes,
and Conventions" documentation [1], size calculations (especially
multiplication) should not be performed in memory allocator (or similar)
function arguments due to the risk of them overflowing. This could lead
to values wrapping around and a smaller allocation being made than the
caller was expecting. Using those allocations could lead to linear
overflows of heap memory and other misbehaviors.

So, use the purpose specific kcalloc() function instead of the argument
size * count in the kzalloc() function.

[1] https://www.kernel.org/doc/html/v5.14/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments

Signed-off-by: Len Baker <len.baker@gmx.com>
Link: https://lore.kernel.org/r/20210904131714.2312-1-len.baker@gmx.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: mt8173: Switch to clk_hw provider APIs

As part of the effort to improve the MediaTek clk drivers, the next step
is to switch from the old 'struct clk' clk prodivder APIs to the new
'struct clk_hw' ones.

The MT8173 clk driver has one clk that is registered directly with the
clk provider APIs, instead of going through the MediaTek clk library.

Switch this instance to use the clk_hw provider API.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220519071610.423372-6-wenst@chromium.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Switch to clk_hw provider APIs

As part of the effort to improve the MediaTek clk drivers, the next step
is to switch from the old 'struct clk' clk prodivder APIs to the new
'struct clk_hw' ones.

In a previous patch, 'struct clk_onecell_data' was replaced with
'struct clk_hw_onecell_data', with (struct clk_hw *)->clk and
__clk_get_hw() bridging the new data structures and old code.

Now switch from the old 'clk_(un)?register*()' APIs to the new
'clk_hw_(un)?register*()' ones. This is done with the coccinelle script
below.

Unfortunately this also leaves clk-mt8173.c with a compile error that
would need a coccinelle script longer than the actual diff to fix. This
last part is fixed up by hand.

    // Fix prototypes
    @@
    identifier F =~ "^mtk_clk_register_";
    @@
    - struct clk *
    + struct clk_hw *
      F(...);

    // Fix calls to mtk_clk_register_<singular>
    @ reg @
    identifier F =~ "^mtk_clk_register_";
    identifier FS =~ "^mtk_clk_register_[a-z_]*s";
    identifier I;
    expression clk_data;
    expression E;
    @@
      FS(...) {
    ...
    - struct clk *I;
    + struct clk_hw *hw;
    ...
    for (...;...;...) {
    ...
    (
    - I
    + hw
    =
    - clk_register_fixed_rate(
    + clk_hw_register_fixed_rate(
    ...
    );
    |
    - I
    + hw
    =
    - clk_register_fixed_factor(
    + clk_hw_register_fixed_factor(
    ...
    );
    |
    - I
    + hw
    =
    - clk_register_divider(
    + clk_hw_register_divider(
    ...
    );
    |
    - I
    + hw
    =
    F(...);
    )
    ...
    if (
    -     IS_ERR(I)
    +     IS_ERR(hw)
       ) {
    pr_err(...,
    -        I
    +        hw
    ,...);
    ...
    }

    - clk_data->hws[E] = __clk_get_hw(I);
    + clk_data->hws[E] = hw;
    }
    ...
      }

    @ depends on reg @
    identifier reg.I;
    @@
      return PTR_ERR(
    - I
    + hw
      );

    // Fix mtk_clk_register_composite to return clk_hw instead of clk
    @@
    identifier I, R;
    expression E;
    @@
    - struct clk *
    + struct clk_hw *
      mtk_clk_register_composite(...) {
    ...
    - struct clk *I;
    + struct clk_hw *hw;
    ...
    - I = clk_register_composite(
    + hw = clk_hw_register_composite(
    ...);
    if (IS_ERR(
    -    I
    +    hw
       )) {
    ...
    R = PTR_ERR(
    -       I
    +       hw
  );
    ...
    }

    return
    - I
    + hw
    ;
    ...
      }

    // Fix other mtk_clk_register_<singular> to return clk_hw instead of clk
    @@
    identifier F =~ "^mtk_clk_register_";
    identifier I, D, C;
    expression E;
    @@
    - struct clk *
    + struct clk_hw *
      F(...) {
    ...
    - struct clk *I;
    + int ret;
    ...
    - I = clk_register(D, E);
    + ret = clk_hw_register(D, E);
    ...
    (
    - if (IS_ERR(I))
    + if (ret) {
    kfree(C);
    + return ERR_PTR(ret);
    + }
    |
    - if (IS_ERR(I))
    + if (ret)
    {
    kfree(C);
    - return I;
    + return ERR_PTR(ret);
    }
    )

    - return I;
    + return E;
      }

    // Fix mtk_clk_unregister_<singular> to take clk_hw instead of clk
    @@
    identifier F =~ "^mtk_clk_unregister_";
    identifier I, I2;
    @@
      static void F(
    -  struct clk *I
    + struct clk_hw *I2
      )
      {
    ...
    - struct clk_hw *I2;
    ...
    - I2 = __clk_get_hw(I);
    ...
    (
    - clk_unregister(I);
    + clk_hw_unregister(I2);
    |
    - clk_unregister_composite(I);
    + clk_hw_unregister_composite(I2);
    )
    ...
      }

    // Fix calls to mtk_clk_unregister_*()
    @@
    identifier F =~ "^mtk_clk_unregister_";
    expression I;
    expression E;
    @@
    - F(I->hws[E]->clk);
    + F(I->hws[E]);

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220519071610.423372-5-wenst@chromium.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Replace 'struct clk' with 'struct clk_hw'

As part of the effort to improve the MediaTek clk drivers, the next step
is to switch from the old 'struct clk' clk prodivder APIs to the new
'struct clk_hw' ones.

Instead of adding new APIs to the MediaTek clk driver library mirroring
the existing ones, moving all drivers to the new APIs, and then removing
the old ones, just migrate everything at the same time. This involves
replacing 'struct clk' with 'struct clk_hw', and 'struct clk_onecell_data'
with 'struct clk_hw_onecell_data', and fixing up all usages.

For now, the clk_register() and co. usage is retained, with __clk_get_hw()
and (struct clk_hw *)->clk used to bridge the difference between the APIs.
These will be replaced in subsequent patches.

Fix up mtk_{alloc,free}_clk_data to use 'struct clk_hw' by hand. Fix up
all other affected call sites with the following coccinelle script.

    // Replace type
    @@
    @@
    - struct clk_onecell_data
    + struct clk_hw_onecell_data

    // Replace of_clk_add_provider() & of_clk_src_simple_get()
    @@
    expression NP, DATA;
    symbol of_clk_src_onecell_get;
    @@
    - of_clk_add_provider(
    + of_clk_add_hw_provider(
    NP,
    - of_clk_src_onecell_get,
    + of_clk_hw_onecell_get,
    DATA
      )

    // Fix register/unregister
    @@
    identifier CD;
    expression E;
    identifier fn =~ "unregister";
    @@
      fn(...,
    -    CD->clks[E]
    +    CD->hws[E]->clk
 ,...
);

    // Fix calls to clk_prepare_enable()
    @@
    identifier CD;
    expression E;
    @@
      clk_prepare_enable(
    -       CD->clks[E]
    +       CD->hws[E]->clk
      );

    // Fix pointer assignment
    @@
    identifier CD;
    identifier CLK;
    expression E;
    @@
    - CD->clks[E]
    + CD->hws[E]
      =
    (
    - CLK
    + __clk_get_hw(CLK)
    |
      ERR_PTR(...)
    )
      ;

    // Fix pointer usage
    @@
    identifier CD;
    expression E;
    @@
    - CD->clks[E]
    + CD->hws[E]

    // Fix mtk_clk_pll_get_base()
    @@
    symbol clk, hw, data;
    @@
      mtk_clk_pll_get_base(
    -         struct clk *clk,
    +         struct clk_hw *hw,
   const struct mtk_pll_data *data
      ) {
    - struct clk_hw *hw = __clk_get_hw(clk);
      ...
      }

    // Fix mtk_clk_pll_get_base() usage
    @@
    identifier CD;
    expression E;
    @@
      mtk_clk_pll_get_base(
    -    CD->clks[E]
    +    CD->hws[E]->clk
 ,...
      );

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220519071610.423372-4-wenst@chromium.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: apmixed: Drop error message from clk_register() failure

mtk_clk_register_ref2usb_tx() prints an error message if clk_register()
fails. It doesn't if kzalloc() fails though. The caller would then tack
on its own error message to handle this.

Also, All other clk registration functions in the MediaTek clk library
leave the error message printing to the bulk registration functions,
while the helpers that register individual clks just return error codes.

Drop the error message that is printed when clk_register() fails in
mtk_clk_register_ref2usb_tx() to make its behavior consistent both
across its failure modes, and with the rest of the driver library.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220519071610.423372-3-wenst@chromium.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Make mtk_clk_register_composite() static

mtk_clk_register_composite() is not used anywhere outside of the file it
is defined.

Make it static.

Fixes: 9741b1a68035 ("clk: mediatek: Add initial common clock support for Mediatek SoCs.")
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220519071610.423372-2-wenst@chromium.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: use en_mask as a pure div_en_mask

We no longer allow en_mask to be a combination of
pll_en_bit and div_en_mask, so remove pll_en_bit(bit0)
from en_mask to make en_mask a pure en_mask that only
used for pll dividers.

This commit continues the work done in commit 7cc4e1bbe300
("clk: mediatek: Fix asymmetrical PLL enable and disable
control") and commit f384c44754b7 ("clk: mediatek:
Add configurable enable control to mtk_pll_data") to
clean up en_mask(bit0) default setting.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Signed-off-by: Mandy Liu <mandyjh.liu@mediatek.com>
Link: https://lore.kernel.org/r/20220513073621.12923-1-mandyjh.liu@mediatek.com
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: update compatible string for MT7986 ethsys

Update compatible string of mt7986 ethsys clock driver to fit the
devicetree bindings document.

Signed-off-by: Sam Shih <sam.shih@mediatek.com>
Link: https://lore.kernel.org/r/20220509090939.845-2-sam.shih@mediatek.com
Fixes: ec97d23c8e22 ("clk: mediatek: add mt7986 clock support")
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: ux500: fix a possible off-by-one in u8500_prcc_reset_base()

Off-by-one will happen when index == ARRAY_SIZE(ur->base).

Fixes: b14cbdfd467d ("clk: ux500: Add driver for the reset portions of PRCC")
Signed-off-by: Hangyu Hua <hbh25y@gmail.com>
Link: https://lore.kernel.org/r/20220518062537.17933-1-hbh25y@gmail.com
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

Merge tag 'clk-imx-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa/linux into clk-imx

Pull i.MX clk driver updates from Abel Vesa:

 - Add 27 MHz phy PLL ref clock
 - Add mcore_booted module parameter to tell kernel M core has already booted
 - Remove snvs clock
 - Add bindings for i.MX8MN GPT
 - Add check for kcalloc
 - Fix for a potential memory leak in __imx_clk_gpr_sync
 - Add DISP2 pixel clock for i.MX8MP
 - Use pm_runtime_resume_and_get to fix pm_runtime_get_sync() usage
 - Add clkout1/2 for i.MX8MP
 - Fix parent clock of ubs_root_clk for i.MX8MP

* tag 'clk-imx-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/abelvesa/linux:
  clk: imx8mp: fix usb_root_clk parent
  clk: imx8mp: add clkout1/2 support
  clk: imx: scu: Use pm_runtime_resume_and_get to fix pm_runtime_get_sync() usage
  clk: imx8mp: Add DISP2 pixel clock
  clk: imx: scu: fix a potential memory leak in __imx_clk_gpr_scu()
  clk: imx: Add check for kcalloc
  clk: imx8mn: add GPT support
  dt-bindings: imx: add clock bindings for i.MX8MN GPT
  clk: imx: Remove the snvs clock
  clk: imx8m: check mcore_booted before register clk
  clk: imx: add mcore_booted module paratemter
  clk: imx8mq: add 27m phy pll ref clock

clk: imx8mp: fix usb_root_clk parent

According to reference mannual CCGR77(usb) sources from hsio_axi, fix
it.

Fixes: 9c140d9926761 ("clk: imx: Add support for i.MX8MP clock driver")
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20220507125430.793287-1-peng.fan@oss.nxp.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>

Merge tag 'sunxi-clk-for-5.19-1' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into clk-allwinner

Pull Allwinner clk driver updates from Jernej Skrabec:

 - Allwinner H6 RTC clock
 - Allwinner H616 32 kHz clock

* tag 'sunxi-clk-for-5.19-1' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
  clk: sunxi-ng: h616: Add PLL derived 32KHz clock
  clk: sunxi-ng: h6-r: Add RTC gate clock

Merge tag 'for-5.19-clk' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux into clk-tegra

Pull Tegra clk driver updates from Thierry Reding:

This contains a boot time optimization for Tegra chips with BPMP and a
switch from .round_rate() to .determine_rate() to take into account any
maximum rate that might have been set.

Other than that this contains a fix for a DFLL regression on Tegra210
and kerneldoc fixups to avoid build warnings.

* tag 'for-5.19-clk' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
  clk: tegra: Update kerneldoc to match prototypes
  clk: tegra: Replace .round_rate() with .determine_rate()
  clk: tegra: Register clocks from root to leaf
  clk: tegra: Add missing reset deassertion

Merge tag 'renesas-clk-for-v5.19-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers into clk-renesas

Pull Renesas clk driver updates from Geert Uytterhoeven:

 - Add support for the R-Car V4H and RZ/V2M SoCs
 - Add the Universal Flash Storage clock on R-Car S4-8
 - Add I2C, SSIF-2 (sound), USB, CANFD, OSTM (timer), WDT, SPI Multi
   I/O Bus, RSPI, TSU (thermal), and ADC clocks and resets on RZ/G2UL
 - Add display clock support on RZ/G2L
 - Miscellaneous fixes and improvements

* tag 'renesas-clk-for-v5.19-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers: (36 commits)
  clk: renesas: r9a09g011: Add eth clock and reset entries
  clk: renesas: Add RZ/V2M support using the rzg2l driver
  clk: renesas: rzg2l: Add support for RZ/V2M reset monitor reg
  clk: renesas: rzg2l: Make use of CLK_MON registers optional
  clk: renesas: rzg2l: Set HIWORD mask for all mux and dividers
  clk: renesas: rzg2l: Add read only versions of the clk macros
  clk: renesas: rzg2l: Move the DEF_MUX array size calc into the macro
  dt-bindings: clock: renesas,rzg2l: Document RZ/V2M SoC
  clk: renesas: r9a07g044: Fix OSTM1 module clock name
  clk: renesas: r9a07g043: Add clock and reset entries for ADC
  clk: renesas: r9a07g043: Add TSU clock and reset entry
  clk: renesas: r9a07g043: Add RSPI clock and reset entries
  clk: renesas: r9a07g043: Add clock and reset entries for SPI Multi I/O Bus Controller
  clk: renesas: r9a07g044: Add DSI clock and reset entries
  clk: renesas: r9a07g044: Add LCDC clock and reset entries
  clk: renesas: r9a07g044: Add M4 Clock support
  clk: renesas: r9a07g044: Add M3 Clock support
  clk: renesas: r9a07g044: Add {M2, M2_DIV2} Clocks support
  clk: renesas: r9a07g044: Add M1 clock support
  clk: renesas: rzg2l: Add DSI divider clk support
  ...

clk: fixed-rate: Remove redundant if statement

(np) is always true when (dev || !np) is false, so just remove
the check.

Signed-off-by: Li Zhengyu <lizhengyu3@huawei.com>
Link: https://lore.kernel.org/r/20220509092102.140520-1-lizhengyu3@huawei.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: sunxi-ng: h616: Add PLL derived 32KHz clock

The RTC section of the H616 manual mentions in a half-sentence the
existence of a clock "32K divided by PLL_PERI(2X)". This is used as
one of the possible inputs for the mux that selects the clock for the
32 KHz fanout pad. On the H616 this is routed to pin PG10, and some
boards use that clock output to compensate for a missing 32KHz crystal.
On the OrangePi Zero2 this is for instance connected to the LPO pin of
the WiFi/BT chip.
The new RTC clock binding requires this clock to be named as one input
clock, so we need to expose this to the DT. In contrast to the D1 SoC
there does not seem to be a gate for this clock, so just use a fixed
divider clock, using a newly assigned clock number.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Link: https://lore.kernel.org/r/20220428230933.15262-3-andre.przywara@arm.com

clk: sunxi-ng: h6-r: Add RTC gate clock

The H6 and H616 feature an (undocumented) bus clock gate for accessing
the RTC registers. This seems to be enabled at reset (or by the BootROM),
so we got away without it so far, but exists regardless.
Since the new RTC clock binding for the H616 requires this "bus" clock
to be specified in the DT, add this to R_CCU clock driver and expose it
on the DT side with a new number.
We do this for both the H6 and H616, but mark it as IGNORE_UNUSED, as we
cannot reference it in any H6 DTs.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Link: https://lore.kernel.org/r/20220428230933.15262-2-andre.przywara@arm.com

clk: tegra: Update kerneldoc to match prototypes

For some DFLL functions, the kerneldoc comments don't match the function
prototype. Fix them up to avoid some warnings at build time.

Signed-off-by: Thierry Reding <treding@nvidia.com>

clk: renesas: r9a09g011: Add eth clock and reset entries

Add ethernet clock/reset entries to CPG driver.

Note that the AXI and CHI clocks are both enabled and disabled using
the same register bit.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220504145454.71287-2-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: Add RZ/V2M support using the rzg2l driver

The Renesas RZ/V2M SoC is very similar to RZ/G2L, though it doesn't have
any CLK_MON registers.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220503115557.53370-11-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Add support for RZ/V2M reset monitor reg

The RZ/V2M doesn't have a matching set of reset monitor regs for each reset
reg like the RZ/G2L. Instead, it has a single CPG_RST_MON reg which has a
single bit per module.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220503115557.53370-10-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Make use of CLK_MON registers optional

The RZ/V2M SoC doesn't use CLK_MON registers, so make them optional.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220503115557.53370-9-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Set HIWORD mask for all mux and dividers

All of the muxes and dividers that can be modified require the HIWORD
flags, so make the macros set them. It won't affect read only muxes and
dividers.
This will make the clock tables a little easier to read, particularly for
new SoCs coming.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220503115557.53370-8-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Add read only versions of the clk macros

This just makes the clk tables easier to read.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Link: https://lore.kernel.org/r/20220503115557.53370-7-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Move the DEF_MUX array size calc into the macro

We only ever use ARRAY_SIZE() to populate the number of parents, so
move this into the macro to always detect it automatically. This
also makes the tables of clocks a little simpler.
Similarly for the DEF_SD_MUX macro.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Link: https://lore.kernel.org/r/20220503115557.53370-6-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

dt-bindings: clock: renesas,rzg2l: Document RZ/V2M SoC

Document the device tree binding for the Renesas RZ/V2M (r9a09g011) SoC.

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20220503115557.53370-4-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

Merge tag 'renesas-r9a09g011-dt-binding-defs-tag' into renesas-clk-for-v5.19

Renesas RZ/V2M DT Binding Definitions

Clock definitions for the Renesas RZ/V2M (R9A09G011) SoC, shared by
driver and DT source files.

clk: renesas: r9a07g044: Fix OSTM1 module clock name

Fix a typo in the name of the "ostm1_pclk" clock.
This change has no run-time impact.

Fixes: 161450134ae9bab3 ("clk: renesas: r9a07g044: Add OSTM clock and reset entries")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/e0eff1f57378ec29d0d3f1a7bdd7e380583f736b.1651494871.git.geert+renesas@glider.be

clk: renesas: r9a07g043: Add clock and reset entries for ADC

Add clock and reset entries for ADC block in CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220501083450.26541-5-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add TSU clock and reset entry

Add TSU clock and reset entry to CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220501083450.26541-4-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add RSPI clock and reset entries

Add RSPI{0,1,2} clock and reset entries to CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220501083450.26541-3-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add clock and reset entries for SPI Multi I/O Bus Controller

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

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220501083450.26541-2-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g044: Add DSI clock and reset entries

Add DSI clock and reset entries to CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-10-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g044: Add LCDC clock and reset entries

Add LCDC clock and reset entries to CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-9-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g044: Add M4 Clock support

Add support for M4 clock which is sourced from pll2_533_div2.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-8-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g044: Add M3 Clock support

Add support for M3 clock which is sourced from DSI divider connected
to PLL5_4 mux.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-7-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g044: Add {M2, M2_DIV2} Clocks support

Add support for {M2, M2_DIV2} clocks which is sourced from pll3_533.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-6-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g044: Add M1 clock support

Add support for M1 clock which is sourced from FOUTPOSTDIV.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-5-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Add DSI divider clk support

M3 clock is sourced from DSI Divider (DSIDIVA * DSIDIVB)

This patch add support for DSI divider clk by combining
DSIDIVA and DSIDIVB.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-4-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Add PLL5_4 clk mux support

Add PLL5_4 clk mux support to select clock from clock
sources FOUTPOSTDIV and FOUT1PH0.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-3-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: rzg2l: Add FOUTPOSTDIV clk support

PLL5 generates FOUTPOSTDIV clk and is sourced by LCDC/DSI modules.
The FOUTPOSTDIV is connected to PLL5_4 MUX. Video clock is sourced
from DSI divider which is connected to PLL5_4 MUX.

This patch adds support for generating FOUTPOSTDIV clk.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220430114156.6260-2-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

dt-bindings: clock: Add r9a09g011 CPG Clock Definitions

Define RZ/V2M (R9A09G011) Clock Pulse Generator module clock outputs
(CPG_CLK_ON* registers), and reset definitions (CPG_RST_* registers)
in Section 48.5 ("Register Description") of the RZ/V2M Hardware User's
Manual (Rev. 1.10, Sep. 2021).

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220503115557.53370-3-phil.edworthy@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: tegra: Replace .round_rate() with .determine_rate()

Replace the .round_rate() callback with .determine_rate() which can
consider max_rate imposed by clk_set_max_rate() while rounding the clock
rate.

Note that if the .determine_rate() callback is defined it will be called
instead of the .round_rate() callback when calling clk_round_rate(). By
using .determine_rate(), the maximum rate returned when calling
clk_round_rate() is now limited by the current max_rate.

Signed-off-by: Rajkumar Kasirajan <rkasirajan@nvidia.com>
[jonathanh@nvidia.com: checkpatch fixes and commit message update]
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

clk: tegra: Register clocks from root to leaf

Current clock initialization causes intermediate registering of orphan
clocks (i.e. a clock without a parent registered). CCF keeps track of
orphan clocks and any time a new clock is registered, it will loop
through the list of orphan and queries if the parent is now available.
This operation triggers one or more clock operations, which are IPCs
with BPMP-FW. Hence, due to the order in which the clocks appear
currently, this causes > 5000 IPC messages to be sent to BPMP-FW during
clock initialization.

Optimize the clock probing by registering clocks hierarchically from
root clock towards leafs.

Signed-off-by: Timo Alho <talho@nvidia.com>
[jonathanh@nvidia.com: checkpatch fixes]
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

clk: tegra: Add missing reset deassertion

Commit 4782c0a5dd88 ("clk: tegra: Don't deassert reset on enabling
clocks") removed deassertion of reset lines when enabling peripheral
clocks. This breaks the initialization of the DFLL driver which relied
on this behaviour.

Fix this problem by adding explicit deassert/assert requests to the
driver. Tested on Google Pixel C.

Cc: stable@vger.kernel.org
Fixes: 4782c0a5dd88 ("clk: tegra: Don't deassert reset on enabling clocks")
Signed-off-by: Diogo Ivo <diogo.ivo@tecnico.ulisboa.pt>
Signed-off-by: Thierry Reding <treding@nvidia.com>

clk: imx8mp: add clkout1/2 support

clkout1 and clkout2 allow to supply clocks from the SoC to the board,
which is used by some board designs to provide reference clocks.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20220427162131.3127303-1-l.stach@pengutronix.de
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>

clk: imx: scu: Use pm_runtime_resume_and_get to fix pm_runtime_get_sync() usage

If the device is already in a runtime PM enabled state
pm_runtime_get_sync() will return 1.

Also, we need to call pm_runtime_put_noidle() when pm_runtime_get_sync()
fails, so use pm_runtime_resume_and_get() instead. this function
will handle this.

Fixes: 78edeb080330 ("clk: imx: scu: add runtime pm support")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Abel Vesa <abel.vesa@nxp.com>
Link: https://lore.kernel.org/r/20220425011117.25093-1-linmq006@gmail.com
Signed-off-by: Abel Vesa <abel.vesa@nxp.com>

clk: renesas: cpg-mssr: Add support for R-Car V4H

Initial CPG support for R-Car V4H (r8a779g0).

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/20220428135058.597586-2-yoshihiro.shimoda.uh@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

Merge tag 'renesas-r8a779g0-dt-binding-defs-tag' into renesas-clk-for-v5.19

Renesas R-Car V4H DT Binding Definitions

Clock and Power Domain definitions for the Renesas R-Car V4H (R8A779G0)
SoC, shared by driver and DT source files.

clk: renesas: rcar-gen4: Add CLK_TYPE_GEN4_PLL4

R-Car V4H (r8a779g0) has PLL4 so that add CLK_TYPE_GEN4_PLL4.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/20220425064201.459633-5-yoshihiro.shimoda.uh@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add WDT clock and reset entries

Add WDT{0,2} clock and reset entries to CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220425095244.156720-7-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add OSTM clock and reset entries

Add OSTM{0,1,2} clock and reset entries to CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220425095244.156720-6-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add clock and reset entries for CANFD

Add clock and reset entries for CANFD in CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220425095244.156720-5-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add USB clocks/resets

Add clock/reset entries for USB PHY control, USB2.0 host and device.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220425095244.156720-4-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add SSIF-2 clock and reset entries

Add SSIF-2{0,1,2,3} clock and reset entries in CPG driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220425095244.156720-3-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a07g043: Add I2C clocks/resets

Add I2C{0,1,2,3} clock and reset entries.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20220425095244.156720-2-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r9a06g032: Fix the RTC hclock description

It needs to be un-gated, but also a reset must be released and an idle
flag should also be disabled.

The driver already supports all these operations, so update the
description of the RTC hclock to fit these requirements.

Fixes: 4c3d88526eba2143 ("clk: renesas: Renesas R9A06G032 clock driver")
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Acked-by: Stephen Boyd <sboyd@kernel.org>
Link: https://lore.kernel.org/r/20220421090016.79517-3-miquel.raynal@bootlin.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: en7523: fix wrong pointer check in en7523_clk_probe()

Check the real return value of devm_platform_ioremap_resource()
in en7523_clk_probe().

Fixes: 1e6273179190 ("clk: en7523: Add clock driver for Airoha EN7523 SoC")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20220426131539.388382-1-yangyingliang@huawei.com
Acked-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 ipesys clock support

Add MT8186 ipesys clock controller which provides clock gate
control for Image Process Engine.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-16-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 mdpsys clock support

Add MT8186 mdpsys clock controller which provides clock gate
control in Multimedia Data Path.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-15-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 camsys clock support

Add MT8186 camsys clock controllers which provide clock gate
control for camera IP blocks.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-14-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 vencsys clock support

Add MT8186 vencsys clock controller which provide clock gate
control for video encoder.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-13-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 vdecsys clock support

Add MT8186 vdec clock controller which provide clock gate
control for video decoder.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-12-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 imgsys clock support

Add MT8186 imgsys clock controllers which provide clock gate
control for image IP blocks.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-11-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 wpesys clock support

Add MT8186 wpesys clock controllers which provide clock gate
control in Wrapping Engine.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-10-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 mmsys clock support

Add MT8186 mmsys clock controller which provides clock gate
control in video system. This is integrated with mtk-mmsys
driver which will populate device by platform_device_register_data
to start mmsys clock driver.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-9-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 mfgsys clock support

Add MT8186 mfg clock controller which provides clock gate
control for GPU.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-8-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 imp i2c wrapper clock support

Add MT8186 imp i2c wrapper clock controllers which provide clock gate
control in i2c IP blocks.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-7-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 apmixedsys clock support

Add MT8186 apmixedsys clock controller which provides Plls
generated from SoC.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-6-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 infrastructure clock support

Add MT8186 infrastructure clock controller which provides
clock gate control for basic IP like pwm, uart, spi and so on.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-5-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 topckgen clock support

Add MT8186 topckgen clock controller which provides muxes, dividers
to handle variety clock selection in other IP blocks.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-4-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mediatek: Add MT8186 mcusys clock support

Add MT8186 mcusys clock controller which provides muxes
to select the clock source of APMCU.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Miles Chen <miles.chen@mediatek.com>
Link: https://lore.kernel.org/r/20220409132251.31725-3-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

dt-bindings: ARM: MediaTek: Add new document bindings of MT8186 clock

This patch adds the new binding documentation for system clock
and functional clock on MediaTek MT8186.

Signed-off-by: Chun-Jie Chen <chun-jie.chen@mediatek.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20220409132251.31725-2-chun-jie.chen@mediatek.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: ux500: Implement the missing CLKOUT clocks

This implements the two missing CLKOUT clocks for the ux500
(well really U8500/DB8500) SoC.

The clocks are initialized using a specific parent and
divider and these are specified in the device tree, see
the separate binding patch.

The implementation is a bit different in that it will only
create the clock in the clock framework if a user appears
in the device tree, rather than it being registered upfront
like most of the other clocks. This is because the clock
needs parameters for source and divider from the consumer
phandle for the clock to be set up properly when the clock
is registered.

There could be more than one user of a CLKOUT clock, but
we have not seen this in practice. If this happens the
framework prints and info and returns the previously
registered clock.

Using the clocks requires also muxing the CLKOUT1 or
CLKOUT2 to the appropriate pad. In practice this is
achived in a pinctrl handle in the DTS node for the device
using the CLKOUT clock, so this muxing is done separately
from the clock itself. Example:

  haptic@49 {
    compatible = "immersion,isa1200";
    reg = <0x49>;
    (...)
    /* clkout1 from ACLK divided by 8 */
    clocks = <&clkout_clk DB8500_CLKOUT_1 DB8500_CLKOUT_SRC_ACLK 8>;
    pinctrl-names = "default";
    pinctrl-0 = <&isa1200_janice_default>;
  };

  isa1200_janice_default: isa1200_janice {
    /* Bring out clkout1 on pin GPIO227 pin AH7 */
    janice_mux {
      function = "clkout";
      groups = "clkout1_a_1";
    };
    janice_cfg1 {
      pins = "GPIO227_AH7";
      ste,config = <&out_lo>;
    };
  (...)

This was tested successfully with the Immersion ISA1200
haptic feedback unit on the Samsung Galaxy S Advance GT-I9070
(Janice) mobile phone.

As the CLKOUT clocks need some undefined fixed rate parent
clocks that are currently missing from the PRCMU clock
implementation, the three simplest are added in this patch:
clk38m_to_clkgen, aclk and sysclk. The only parent not yet
available in the implementation is clk009, which is a kind
of special muxed and divided clock which isn't even
implemented in the vendor clock driver.

Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20220414221751.323525-6-linus.walleij@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: ux500: Rewrite PRCMU clocks to use clk_hw_*

This rewrites all the u8500 PRCMU clocks and helper functions to
handle clocks using struct clk_hw rather than struct clk, as is
normal for modern clock drivers.

Use clk_hw_register(), of_clk_add_hw_provider() and stack all the
clocks into a compile-time dynamic array of
struct clk_hw_onecell_data.

Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20220414221751.323525-5-linus.walleij@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: ux500: Drop .is_prepared state from PRCMU clocks

The core already keeps a software prepare count. Drop this
custom software prepare count.

The only semantic difference is that if disabling the clock
fails, .is_prepared() will still return 1, but this will only
serve to confuse the framework, the error message is
sufficient.

Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20220414221751.323525-4-linus.walleij@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: ux500: Drop .is_enabled state from PRCMU clocks

The core already keeps a software enable count. Drop this
custom software enable count.

Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20220414221751.323525-3-linus.walleij@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

dt-bindings: clock: u8500: Add clkout clock bindings

This adds device tree bindings for the externally routed clocks
CLKOUT1 and CLKOUT2 clocks found in the DB8500.

Cc: devicetree@vger.kernel.org
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20220414221751.323525-2-linus.walleij@linaro.org
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: mux: remove redundant initialization of variable width

Variable width is being ininitialized with a value that is never read.
The ininitializtion is redundant and can be removed. Move the variable
to the scope it is required.

Cleans up cppcheck warning:
Variable 'width' is assigned a value that is never used.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20220424182227.1364966-1-colin.i.king@gmail.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

dt-bindings: clock: renesas,cpg-mssr: Document r8a779g0

Add binding documentation for the R-Car V4H (R8A779G0) Clock Pulse
Generator.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20220420084255.375700-7-yoshihiro.shimoda.uh@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

clk: renesas: r8a779f0: Add UFS clock

Add the module clock used by the UFS host controller on the Renesas
R-Car S4-8 (R8A779F0) SoC.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/20220411124932.3765571-1-yoshihiro.shimoda.uh@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

dt-bindings: clock: Add r8a779g0 CPG Core Clock Definitions

Add all Clock Pulse Generator Core Clock Outputs for the Renesas
R-Car V4H (R8A779G0) SoC.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/20220425064201.459633-3-yoshihiro.shimoda.uh@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

dt-bindings: power: Add r8a779g0 SYSC power domain definitions

Add power domain indices for R-Car V4H (r8a779g0).

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20220425064201.459633-2-yoshihiro.shimoda.uh@renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

Merge tag 'renesas-clk-for-v5.19-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers into clk-renesas

Pull Renesas clk driver updates from Geert Uytterhoeven:

 - Add support for the new RZ/G2UL SoC
 - Add RPC (QSPI/HyperFlash) clocks on R-Car E3 and D3
 - Miscellaneous fixes and improvements

* tag 'renesas-clk-for-v5.19-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers:
  clk: renesas: r9a07g043: Add SDHI clock and reset entries
  clk: renesas: r9a07g043: Add GbEthernet clock/reset
  clk: renesas: r9a07g043: Add ethernet clock sources
  clk: renesas: r9a07g043: Add GPIO clock and reset entries
  clk: renesas: Add support for RZ/G2UL SoC
  clk: renesas: Move RPC core clocks
  clk: renesas: rzg2l: Simplify multiplication/shift logic
  dt-bindings: clock: Add R9A07G043 CPG Clock and Reset Definitions
  clk: renesas: r8a77995: Add RPC clocks
  clk: renesas: r8a77990: Add RPC clocks
  dt-bindings: clock: renesas: Document RZ/G2UL SoC
  clk: renesas: rzg2l: Remove unused notifiers

clk: renesas-pcie: use simple i2c probe function

The i2c probe function here doesn't use the id information provided in
its second argument, so the single-parameter i2c probe function
("probe_new") can be used instead.

This avoids scanning the identifier tables during probes.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-11-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: si570: use i2c_match_id and simple i2c probe

As part of the ongoing i2c transition to the simple probe
("probe_new"), this patch uses i2c_match_id to retrieve the
driver_data for the probed device. The id parameter is thus no longer
necessary and the simple probe can be used instead.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-10-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: si544: use i2c_match_id and simple i2c probe

As part of the ongoing i2c transition to the simple probe
("probe_new"), this patch uses i2c_match_id to retrieve the
driver_data for the probed device. The id parameter is thus no longer
necessary and the simple probe can be used instead.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-9-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: si5351: use i2c_match_id and simple i2c probe

As part of the ongoing i2c transition to the simple probe
("probe_new"), this patch uses i2c_match_id to retrieve the
driver_data for the probed device. The id parameter is thus no longer
necessary and the simple probe can be used instead.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-8-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: si5341: use simple i2c probe function

The i2c probe function here doesn't use the id information provided in
its second argument, so the single-parameter i2c probe function
("probe_new") can be used instead.

This avoids scanning the identifier tables during probes.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-7-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: si514: use simple i2c probe function

The i2c probe function here doesn't use the id information provided in
its second argument, so the single-parameter i2c probe function
("probe_new") can be used instead.

This avoids scanning the identifier tables during probes.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-6-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: max9485: use simple i2c probe function

The i2c probe function here doesn't use the id information provided in
its second argument, so the single-parameter i2c probe function
("probe_new") can be used instead.

This avoids scanning the identifier tables during probes.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-5-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: cs2000-cp: use simple i2c probe function

The i2c probe function here doesn't use the id information provided in
its second argument, so the single-parameter i2c probe function
("probe_new") can be used instead.

This avoids scanning the identifier tables during probes.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-4-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: cdce925: use i2c_match_id and simple i2c probe

As part of the ongoing i2c transition to the simple probe
("probe_new"), this patch uses i2c_match_id to retrieve the
driver_data for the probed device. The id parameter is thus no longer
necessary and the simple probe can be used instead.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-3-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: cdce706: use simple i2c probe function

The i2c probe function here doesn't use the id information provided in
its second argument, so the single-parameter i2c probe function
("probe_new") can be used instead.

This avoids scanning the identifier tables during probes.

Signed-off-by: Stephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20220407151831.2371706-2-steve@sk2.org
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

clk: en7523: Add clock driver for Airoha EN7523 SoC

This driver only registers fixed rate clocks, since the clocks are fully
initialized by the boot loader and should not be changed later, according
to Airoha.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://lore.kernel.org/r/20220314084409.84394-3-nbd@nbd.name
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>

dt-bindings: Add en7523-scu device tree binding documentation

Adds device tree binding documentation for clocks in the EN7523 SOC.

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://lore.kernel.org/r/20220314084409.84394-2-nbd@nbd.name
Signed-off-by: Stephen Boyd <sboyd@kernel.org>