linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: sylvester.nawrocki@gmail.com (Sylwester Nawrocki)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v7] clk: add MOXA ART SoCs clock driver
Date: Fri, 01 Nov 2013 19:13:50 +0100	[thread overview]
Message-ID: <5273EF5E.7050402@gmail.com> (raw)
In-Reply-To: <1381330482-13846-1-git-send-email-jonas.jensen@gmail.com>

Hi Jonas,

On 10/09/2013 04:54 PM, Jonas Jensen wrote:
> This patch adds MOXA ART SoCs clock driver support.
>
> Signed-off-by: Jonas Jensen<jonas.jensen@gmail.com>
> ---
>
> Notes:
>      Since last version, Adam Jaremko has been helping providing info on
>      the clock registers and the SoC in general. He's been looking at
>      bootloader sources in particular, which seem to do the same things
>      with slight variations.
>
>      Two separate clocks are now registered: clk_pll and clk_apb
>
>      clk_apb provides the same 48MHz frequency as before but the value is
>      calculated from clk_pll (192MHz) which in turn is calculated from a
>      fixed rate reference (12MHz).
>
>      Device drivers (watchdog, clocksource, MMC) should generally be
>      interested in reading apb_clk, not clk_pll.
>
>      clk_pll is obtained by multiplying the content of 0x98100030 (16)
>      with the reference.
>
>      The reference represent the PCB oscillator crystal, on UC-7112-LX
>      this has part number "12.000 KCC 20GT".
>
>      apb_clk is obtained by getting the rate of clk_pll and dividing it.
>
>      I don't think the parent clock can change rate, not without modifying
>      the hardware.
>
>      Changes since v6:
>
>      1. all rates calculated from a fixed rate reference clock
>      2. register two separate clocks, clk_pll and clk_apb
>
>      Applies to next-20130927
>
>   .../bindings/clock/moxa,moxart-clock.txt           |  43 ++++++++
>   drivers/clk/Makefile                               |   1 +
>   drivers/clk/clk-moxart.c                           | 109 +++++++++++++++++++++
>   3 files changed, 153 insertions(+)
>   create mode 100644 Documentation/devicetree/bindings/clock/moxa,moxart-clock.txt
>   create mode 100644 drivers/clk/clk-moxart.c
>
> diff --git a/Documentation/devicetree/bindings/clock/moxa,moxart-clock.txt b/Documentation/devicetree/bindings/clock/moxa,moxart-clock.txt
> new file mode 100644
> index 0000000..61fd327
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/clock/moxa,moxart-clock.txt
> @@ -0,0 +1,43 @@
> +Device Tree Clock bindings for arch-moxart
> +
> +This binding uses the common clock binding[1].
> +
> +[1] Documentation/devicetree/bindings/clock/clock-bindings.txt
> +
> +MOXA ART SoCs allow to determine PLL output and APB frequencies
> +by reading registers holding multiplier and divisor information.
> +
> +Required properties:
> +- compatible : Must be "moxa,moxart-pll-clock" or "moxa,moxart-apb-clock"
> +- #clock-cells : Should be 0
> +- reg : Should contain registers location and length
> +- clocks : Should contain phandle to parent clock
> +
> +Optional properties for "moxa,moxart-pll-clock":
> +- multiplier-reg : Should contain register offset
> +- multiplier-mask : Should contain register mask
> +- multiplier-shift : Should contain number of places the register is shifted to the right

Are those properties actually used ? Couldn't you derive those values
from compatible string ? Also it should normally be said explicitly
which register name we refer to, as specified in the datasheet.
However IIRC describing hardware to register bits level in the device
tree is discouraged.

> +- clock-output-names : Should contain clock name
> +
> +Optional properties for "moxa,moxart-apb-clock":
> +- divisor-reg : Should contain register offset
> +- divisor-mask : Should contain register mask
> +- divisor-shift : Should contain number of places the register is shifted to the right

Ditto.

> +- clock-output-names : Should contain clock name
> +
> +
> +For example:
> +
> +	clk_pll: clk_pll0 at 98100000 {
> +		compatible = "moxa,moxart-pll-clock";
> +		#clock-cells =<0>;
> +		reg =<0x98100000 0x34>;
> +		clocks =<&ref12>;
> +	};
> +
> +	clk_apb: clk_apb at 98100000 {
> +		compatible = "moxa,moxart-apb-clock";
> +		#clock-cells =<0>;
> +		reg =<0x98100000 0x34>;
> +		clocks =<&pll0>;
> +	};
> diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
> index 7b11106..4c8d857 100644
> --- a/drivers/clk/Makefile
> +++ b/drivers/clk/Makefile
> @@ -11,6 +11,7 @@ obj-$(CONFIG_COMMON_CLK)	+= clk-composite.o
>
>   # SoCs specific
>   obj-$(CONFIG_ARCH_BCM2835)	+= clk-bcm2835.o
> +obj-$(CONFIG_ARCH_MOXART)	+= clk-moxart.o
>   obj-$(CONFIG_ARCH_NOMADIK)	+= clk-nomadik.o
>   obj-$(CONFIG_ARCH_HIGHBANK)	+= clk-highbank.o
>   obj-$(CONFIG_ARCH_NSPIRE)	+= clk-nspire.o
> diff --git a/drivers/clk/clk-moxart.c b/drivers/clk/clk-moxart.c
> new file mode 100644
> index 0000000..18b337a
> --- /dev/null
> +++ b/drivers/clk/clk-moxart.c
> @@ -0,0 +1,109 @@
> +/*
> + * MOXA ART SoCs clock driver.
> + *
> + * Copyright (C) 2013 Jonas Jensen
> + *
> + * Jonas Jensen<jonas.jensen@gmail.com>
> + *
> + * This file is licensed under the terms of the GNU General Public
> + * License version 2.  This program is licensed "as is" without any
> + * warranty of any kind, whether express or implied.
> + */
> +
> +#include<linux/clk-provider.h>
> +#include<linux/io.h>
> +#include<linux/of_address.h>
> +#include<linux/clkdev.h>
> +
> +void __init moxart_of_pll_clk_init(struct device_node *node)
> +{
> +	static void __iomem *base;
> +	struct clk *clk, *ref_clk;
> +	unsigned long rate;
> +	unsigned int mul, reg_offset = 0x30, reg_mask = 0x3f, reg_shift = 3;
> +	const char *name = node->name;
> +
> +	of_property_read_u32(node, "multiplier-reg",&reg_offset);
> +	of_property_read_u32(node, "multiplier-mask",&reg_mask);
> +	of_property_read_u32(node, "multiplier-shift",&reg_shift);
> +	of_property_read_string(node, "clock-output-names",&name);
> +
> +	base = of_iomap(node, 0);
> +	if (!base) {
> +		pr_err("%s: of_iomap failed\n", node->full_name);
> +		return;
> +	}
> +
> +	mul = (readl(base + reg_offset)>>  reg_shift)&  reg_mask;
> +	iounmap(base);
> +
> +	ref_clk = of_clk_get(node, 0);
> +	if (IS_ERR(ref_clk)) {
> +		pr_err("%s: of_clk_get failed\n", node->full_name);
> +		return;
> +	}
> +
> +	rate = (mul * clk_get_rate(ref_clk));

Parentheses seem redundant.

> +	clk = clk_register_fixed_rate(NULL, name, NULL, CLK_IS_ROOT, rate);
> +	clk_register_clkdev(clk, NULL, name);

You may want to check return value of this function, before actually
adding the clock provider.

> +	of_clk_add_provider(node, of_clk_src_simple_get, clk);
> +}
> +CLK_OF_DECLARE(moxart_pll_clock, "moxa,moxart-pll-clock",
> +	       moxart_of_pll_clk_init);
> +
> +void __init moxart_of_apb_clk_init(struct device_node *node)
> +{
> +	static void __iomem *base;
> +	struct clk *clk, *pll_clk;
> +	unsigned long rate;
> +	unsigned int div, val, reg_offset = 0xc, reg_mask = 0x7, reg_shift = 4;
> +	const char *name = node->name;
> +
> +	of_property_read_u32(node, "divisor-reg",&reg_offset);
> +	of_property_read_u32(node, "divisor-mask",&reg_mask);
> +	of_property_read_u32(node, "divisor-shift",&reg_shift);
> +	of_property_read_string(node, "clock-output-names",&name);
> +
> +	base = of_iomap(node, 0);
> +	if (!base) {
> +		pr_err("%s: of_iomap failed\n", node->full_name);
> +		return;
> +	}
> +
> +	val = (readl(base + reg_offset)>>  reg_shift)&  reg_mask;
> +	iounmap(base);
> +
> +	switch (val) {
> +	case 1:
> +		div = 3;
> +		break;
> +	case 2:
> +		div = 4;
> +		break;
> +	case 3:
> +		div = 6;
> +		break;
> +	case 4:
> +		div = 8;
> +		break;
> +	default:
> +		div = 2;
> +		break;
> +	}
> +
> +	pll_clk = of_clk_get(node, 0);
> +	if (IS_ERR(pll_clk)) {
> +		pr_err("%s: of_clk_get failed\n", node->full_name);
> +		return;
> +	}
> +
> +	rate = clk_get_rate(pll_clk) / (div * 2);
> +
> +	clk = clk_register_fixed_rate(NULL, name, NULL, CLK_IS_ROOT, rate);
> +	clk_register_clkdev(clk, NULL, name);

Ditto.

> +	of_clk_add_provider(node, of_clk_src_simple_get, clk);
> +}

Regards,
Sylwester

  reply	other threads:[~2013-11-01 18:13 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-06-27 14:03 [PATCH] clk: add MOXA ART SoCs clock driver Jonas Jensen
2013-07-04 13:08 ` [PATCH v2] " Jonas Jensen
2013-07-17 13:23   ` [PATCH v3] " Jonas Jensen
2013-07-18  9:50     ` Mark Rutland
2013-07-18 10:36       ` Jonas Jensen
2013-07-18 11:02         ` Mark Rutland
2013-07-18 11:55           ` Jonas Jensen
2013-07-18 13:56             ` Mark Rutland
2013-07-18 14:25               ` Jonas Jensen
2013-07-19  8:07     ` [PATCH v4] " Jonas Jensen
2013-07-19  8:17       ` [PATCH v5] " Jonas Jensen
2013-07-22  9:21         ` Mark Rutland
2013-07-23  8:09           ` Tomasz Figa
2013-07-26 22:32             ` Mike Turquette
2013-07-29  9:44         ` [PATCH v6] " Jonas Jensen
2013-10-07  4:47           ` Mike Turquette
2013-10-09 14:54           ` [PATCH v7] " Jonas Jensen
2013-11-01 18:13             ` Sylwester Nawrocki [this message]
2013-12-09 15:16             ` [PATCH v8] " Jonas Jensen
2014-01-17 15:03               ` [PATCH v9] " Jonas Jensen
2014-01-17 15:17                 ` Sudeep Holla
2014-01-21 12:44                 ` [PATCH v10] " Jonas Jensen
2014-01-27 10:20                   ` Mark Rutland
2014-01-28 11:09                   ` [PATCH v11] " Jonas Jensen
2014-03-13 20:27                     ` Mike Turquette

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5273EF5E.7050402@gmail.com \
    --to=sylvester.nawrocki@gmail.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).