devicetree.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Philipp Zabel <p.zabel-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
To: Hauke Mehrtens <hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
Cc: ralf-6z/3iImG2C8G8FEW9MqTrA@public.gmane.org,
	linux-mips-6z/3iImG2C8G8FEW9MqTrA@public.gmane.org,
	linux-mtd-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
	linux-watchdog-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	martin.blumenstingl-gM/Ye1E23mwN+BqQ9rBEUg@public.gmane.org,
	john-Pj+rj9U5foFAfugRpC6u6w@public.gmane.org,
	linux-spi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	hauke.mehrtens-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
	robh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org
Subject: Re: [PATCH v2 09/15] reset: Add a reset controller driver for the Lantiq XWAY based SoCs
Date: Mon, 22 May 2017 11:33:33 +0200	[thread overview]
Message-ID: <1495445613.3558.67.camel@pengutronix.de> (raw)
In-Reply-To: <20170521130918.27446-10-hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>

Hi Hauke,

thank you for the patch. I have a few questions and comments below:

On Sun, 2017-05-21 at 15:09 +0200, Hauke Mehrtens wrote:
> From: Martin Blumenstingl <martin.blumenstingl-gM/Ye1E23mwN+BqQ9rBEUg@public.gmane.org>
> 
> The reset controllers (on xRX200 and newer SoCs have two of them) are
> provided by the RCU module. This was initially implemented as a simple
> reset controller. However, the RCU module provides more functionality
> (ethernet GPHYs, USB PHY, etc.), which makes it a MFD device.
> The old reset controller driver implementation from
> arch/mips/lantiq/xway/reset.c did not honor this fact.

Does this driver replace arch/mips/lantiq/xway/reset.c?

> For some devices the request and the status bits are different.
> 
> Signed-off-by: Hauke Mehrtens <hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
> Cc: Philipp Zabel <p.zabel-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
> ---
>  .../devicetree/bindings/reset/lantiq,rcu-reset.txt |  30 +++
>  drivers/reset/Kconfig                              |   6 +
>  drivers/reset/Makefile                             |   1 +
>  drivers/reset/reset-lantiq-rcu.c                   | 201 +++++++++++++++++++++
>  4 files changed, 238 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/reset/lantiq,rcu-reset.txt
>  create mode 100644 drivers/reset/reset-lantiq-rcu.c
> 
> diff --git a/Documentation/devicetree/bindings/reset/lantiq,rcu-reset.txt b/Documentation/devicetree/bindings/reset/lantiq,rcu-reset.txt
> new file mode 100644
> index 000000000000..ac57c4a1b191
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/reset/lantiq,rcu-reset.txt
> @@ -0,0 +1,30 @@
> +Lantiq XWAY SoC RCU reset controller binding
> +============================================
> +
> +This binding describes a reset-controller found on the RCU module on Lantiq
> +XWAY SoCs.
> +
> +
> +-------------------------------------------------------------------------------
> +Required properties (controller (parent) node):
> +- compatible		: Should be "lantiq,rcu-reset"
> +- lantiq,rcu-syscon	: A phandle to the RCU syscon, the reset register
> +			  offset and the status register offset.
> +- #reset-cells		: Specifies the number of cells needed to encode the
> +			  reset line, should be 2.
> +			  The first cell takes the reset set bit and the
> +			  second cell takes the status bit.
> +
> +-------------------------------------------------------------------------------
> +Example for the reset-controllers on the xRX200 SoCs:
> +	rcu_reset0: rcu_reset {

I'd prefer
	rcu_reset0: reset-controller {

> +		compatible = "lantiq,rcu-reset";
> +		lantiq,rcu-syscon = <&rcu0 0x10 0x14>;
> +		#reset-cells = <2>;
> +	};
> +
> +	rcu_reset1: rcu_reset {
> +		compatible = "lantiq,rcu-reset";
> +		lantiq,rcu-syscon = <&rcu0 0x48 0x24>;
> +		#reset-cells = <2>;
> +	};

I think these should be children of the &rcu0 node. Then you could use
the reg property to specify the registers.

Also, have you considered using a binding like the ti-syscon-reset? 
That could remove the need for a rcu_reset node per register pair
altogether, but might not make sense if the reset registers are densely
populated.

> diff --git a/drivers/reset/Kconfig b/drivers/reset/Kconfig
> index d21c07ccc94e..632e82155590 100644
> --- a/drivers/reset/Kconfig
> +++ b/drivers/reset/Kconfig
> @@ -41,6 +41,12 @@ config RESET_IMX7
>  	help
>  	  This enables the reset controller driver for i.MX7 SoCs.
>  
> +config RESET_LANTIQ_RCU
> +	bool "Lantiq XWAY Reset Driver" if COMPILE_TEST
> +	default SOC_TYPE_XWAY
> +	help
> +	  This enables the reset controller driver for Lantiq / Intel XWAY SoCs.
> +
>  config RESET_LPC18XX
>  	bool "LPC18xx/43xx Reset Driver" if COMPILE_TEST
>  	default ARCH_LPC18XX
> diff --git a/drivers/reset/Makefile b/drivers/reset/Makefile
> index 02a74db94339..b19e77847b98 100644
> --- a/drivers/reset/Makefile
> +++ b/drivers/reset/Makefile
> @@ -6,6 +6,7 @@ obj-$(CONFIG_RESET_A10SR) += reset-a10sr.o
>  obj-$(CONFIG_RESET_ATH79) += reset-ath79.o
>  obj-$(CONFIG_RESET_BERLIN) += reset-berlin.o
>  obj-$(CONFIG_RESET_IMX7) += reset-imx7.o
> +obj-$(CONFIG_RESET_LANTIQ_RCU) += reset-lantiq-rcu.o
>  obj-$(CONFIG_RESET_LPC18XX) += reset-lpc18xx.o
>  obj-$(CONFIG_RESET_MESON) += reset-meson.o
>  obj-$(CONFIG_RESET_OXNAS) += reset-oxnas.o
> diff --git a/drivers/reset/reset-lantiq-rcu.c b/drivers/reset/reset-lantiq-rcu.c
> new file mode 100644
> index 000000000000..f24191beed56
> --- /dev/null
> +++ b/drivers/reset/reset-lantiq-rcu.c
> @@ -0,0 +1,201 @@
> +/*
> + *  This program is free software; you can redistribute it and/or modify it
> + *  under the terms of the GNU General Public License version 2 as published
> + *  by the Free Software Foundation.
> + *
> + *  Copyright (C) 2010 John Crispin <blogic-p3rKhJxN3npAfugRpC6u6w@public.gmane.org>
> + *  Copyright (C) 2013-2015 Lantiq Beteiligungs-GmbH & Co.KG
> + *  Copyright (C) 2016 Martin Blumenstingl <martin.blumenstingl-gM/Ye1E23mwN+BqQ9rBEUg@public.gmane.org>
> + *  Copyright (C) 2017 Hauke Mehrtens <hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
> + */
> +
> +#include <linux/mfd/syscon.h>
> +#include <linux/module.h>
> +#include <linux/regmap.h>
> +#include <linux/reset-controller.h>
> +#include <linux/of_platform.h>
> +
> +#define LANTIQ_RCU_RESET_TIMEOUT	1000000
> +
> +struct lantiq_rcu_reset_priv {
> +	struct reset_controller_dev rcdev;
> +	struct device *dev;
> +	struct regmap *regmap;
> +	u32 reset_offset;
> +	u32 status_offset;
> +};
> +
> +static struct lantiq_rcu_reset_priv *to_lantiq_rcu_reset_priv(
> +	struct reset_controller_dev *rcdev)
> +{
> +	return container_of(rcdev, struct lantiq_rcu_reset_priv, rcdev);
> +}
> +
> +static int lantiq_rcu_reset_status(struct reset_controller_dev *rcdev,
> +			     unsigned long id)

Please align with the opening parenthesis above, or use consistent
indentation for broken lines, also in the other cases below.

> +{
> +	struct lantiq_rcu_reset_priv *priv = to_lantiq_rcu_reset_priv(rcdev);
> +	unsigned int status = (id >> 8) & 0x1f;
> +	u32 val;
> +	int ret;
> +
> +	if (status >= rcdev->nr_resets)
> +		return -EINVAL;

This should be checked in lantiq_rcu_reset_xlate instead.

> +	ret = regmap_read(priv->regmap, priv->status_offset, &val);
> +	if (ret)
> +		return ret;
> +
> +	return !!(val & BIT(status));
> +}
> +
> +static int lantiq_rcu_reset_update(struct reset_controller_dev *rcdev,
> +				   unsigned long id, bool assert)
> +{
> +	struct lantiq_rcu_reset_priv *priv = to_lantiq_rcu_reset_priv(rcdev);
> +	int ret, retry = LANTIQ_RCU_RESET_TIMEOUT;
> +	unsigned int set = id & 0x1f;
> +	u32 val;
> +
> +	if (set >= rcdev->nr_resets)
> +		return -EINVAL;

This should be checked in lantiq_rcu_reset_xlate instead.

> +	if (assert)
> +		val = BIT(set);
> +	else
> +		val = 0;
> +
> +	ret = regmap_update_bits(priv->regmap, priv->reset_offset, BIT(set),
> +				 val);
> +	if (ret) {
> +		dev_err(priv->dev, "Failed to set reset bit %u\n", set);
> +		return ret;
> +	}
> +
> +	do {} while (--retry && lantiq_rcu_reset_status(rcdev, id) != assert);

This ignores errors returned by lantiq_rcu_reset_status. Consider using
regmap_read_poll_timeout() instead of open-coding the timeout.

> +	if (!retry) {
> +		dev_err(priv->dev, "Failed to %s bit %u\n",
> +			assert ? "assert" : "deassert", set);
> +		return -EIO;
> +	}
> +
> +	return 0;
> +}
> +
> +static int lantiq_rcu_reset_assert(struct reset_controller_dev *rcdev,
> +			     unsigned long id)
> +{
> +	return lantiq_rcu_reset_update(rcdev, id, true);
> +}
> +
> +static int lantiq_rcu_reset_deassert(struct reset_controller_dev *rcdev,
> +			       unsigned long id)
> +{
> +	return lantiq_rcu_reset_update(rcdev, id, false);
> +}
> +
> +static int lantiq_rcu_reset_reset(struct reset_controller_dev *rcdev,
> +			    unsigned long id)
> +{
> +	int ret;
> +
> +	ret = lantiq_rcu_reset_assert(rcdev, id);
> +	if (ret)
> +		return ret;
> +
> +	return lantiq_rcu_reset_deassert(rcdev, id);
> +}
> +
> +static struct reset_control_ops lantiq_rcu_reset_ops = {

reset_control_ops should be const.

> +	.assert = lantiq_rcu_reset_assert,
> +	.deassert = lantiq_rcu_reset_deassert,
> +	.status = lantiq_rcu_reset_status,
> +	.reset	= lantiq_rcu_reset_reset,
> +};
> +
> +static int lantiq_rcu_reset_of_probe(struct platform_device *pdev,
> +			       struct lantiq_rcu_reset_priv *priv)
> +{
> +	struct device_node *np = pdev->dev.of_node;
> +
> +	priv->regmap = syscon_regmap_lookup_by_phandle(np,
> +							"lantiq,rcu-syscon");

This could be
	priv->regmap = syscon_node_to_regmap(np->parent);
if the reset controller device node is a child of the MFD device node.

> +	if (IS_ERR(priv->regmap)) {
> +		dev_err(&pdev->dev, "Failed to lookup RCU regmap\n");
> +		return PTR_ERR(priv->regmap);
> +	}
> +
> +	if (of_property_read_u32_index(np, "lantiq,rcu-syscon", 1,
> +		&priv->reset_offset)) {
> +		dev_err(&pdev->dev, "Failed to get RCU reset offset\n");
> +		return -EINVAL;
> +	}
> +
> +	if (of_property_read_u32_index(np, "lantiq,rcu-syscon", 2,
> +		&priv->status_offset)) {
> +		dev_err(&pdev->dev, "Failed to get RCU status offset\n");
> +		return -EINVAL;
> +	}
> +
> +	return 0;
> +}
> +
> +static int lantiq_rcu_reset_xlate(struct reset_controller_dev *rcdev,
> +				  const struct of_phandle_args *reset_spec)
> +{
> +	unsigned int status, set;
> +
> +	set = reset_spec->args[0];
> +	status = reset_spec->args[1];

Check that set and status are valid here.

> +	return (status << 8) | set;
> +}
> +
> +static int lantiq_rcu_reset_probe(struct platform_device *pdev)
> +{
> +	struct lantiq_rcu_reset_priv *priv;
> +	int err;
> +
> +	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
> +	if (!priv)
> +		return -ENOMEM;
> +
> +	priv->dev = &pdev->dev;
> +	platform_set_drvdata(pdev, priv);
> +
> +	err = lantiq_rcu_reset_of_probe(pdev, priv);
> +	if (err)
> +		return err;
> +
> +	priv->rcdev.ops = &lantiq_rcu_reset_ops;
> +	priv->rcdev.owner = THIS_MODULE;
> +	priv->rcdev.of_node = pdev->dev.of_node;
> +	priv->rcdev.nr_resets = 32;
> +	priv->rcdev.of_xlate = lantiq_rcu_reset_xlate;
> +	priv->rcdev.of_reset_n_cells = 2;
> +
> +	err = reset_controller_register(&priv->rcdev);
> +	if (err)
> +		return err;
> +
> +	return 0;

You can just
	return reset_controller_register(&priv->rcdev);
here.

> +}
> +
> +static const struct of_device_id lantiq_rcu_reset_dt_ids[] = {
> +	{ .compatible = "lantiq,rcu-reset", },
> +	{ },
> +};
> +MODULE_DEVICE_TABLE(of, lantiq_rcu_reset_dt_ids);
> +
> +static struct platform_driver lantiq_rcu_reset_driver = {
> +	.probe	= lantiq_rcu_reset_probe,
> +	.driver = {
> +		.name		= "lantiq-rcu-reset",
> +		.of_match_table	= lantiq_rcu_reset_dt_ids,
> +	},
> +};
> +module_platform_driver(lantiq_rcu_reset_driver);
> +
> +MODULE_AUTHOR("Martin Blumenstingl <martin.blumenstingl-gM/Ye1E23mwN+BqQ9rBEUg@public.gmane.org>");
> +MODULE_DESCRIPTION("Lantiq XWAY RCU Reset Controller Driver");
> +MODULE_LICENSE("GPL");

regards
Philipp

--
To unsubscribe from this list: send the line "unsubscribe linux-watchdog" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2017-05-22  9:33 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-21 13:09 [PATCH v2 00/15] MIPS: lantiq: handle RCU register by separate drivers Hauke Mehrtens
     [not found] ` <20170521130918.27446-1-hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
2017-05-21 13:09   ` [PATCH v2 01/15] MIPS: lantiq: Use of_platform_populate instead of __dt_register_buses Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 02/15] mtd: lantiq-flash: drop check of boot select Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 03/15] mtd: spi-falcon: " Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 04/15] watchdog: lantiq: access boot cause register through regmap Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 05/15] MIPS: lantiq: Enable MFD_SYSCON to be able to use it for the RCU MFD Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 06/15] Documentation: DT: MIPS: lantiq: Add docs for the RCU bindings Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 07/15] MIPS: lantiq: Convert the xbar driver to a platform_driver Hauke Mehrtens
     [not found]     ` <20170521130918.27446-8-hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
2017-05-22  5:30       ` John Crispin
     [not found]         ` <71c88b1c-3ed1-8f30-1dbd-dbd4c7bb5cdc-Pj+rj9U5foFAfugRpC6u6w@public.gmane.org>
2017-05-23 20:56           ` Hauke Mehrtens
2017-05-22  6:05       ` Andy Shevchenko
     [not found]         ` <CAHp75VeQgekiWc+YxP5sDFBB4fvmRs1=heFcdYS6mAgqPACW7g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-05-25 18:22           ` Hauke Mehrtens
     [not found]             ` <aa0f9259-b96f-f072-93d4-a42e05fadc82-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
2017-05-25 18:54               ` Hauke Mehrtens
2017-05-30 23:11       ` Rob Herring
2017-05-21 13:09   ` [PATCH v2 08/15] MIPS: lantiq: remove ltq_reset_cause() and ltq_boot_select() Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 09/15] reset: Add a reset controller driver for the Lantiq XWAY based SoCs Hauke Mehrtens
     [not found]     ` <20170521130918.27446-10-hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
2017-05-22  9:33       ` Philipp Zabel [this message]
     [not found]         ` <1495445613.3558.67.camel-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
2017-05-23 21:25           ` Hauke Mehrtens
     [not found]             ` <2f809dc7-53c6-ebf9-53c1-466bf34e39db-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
2017-05-24 13:10               ` Philipp Zabel
2017-05-21 13:09   ` [PATCH v2 10/15] MIPS: lantiq: remove old reset controller implementation Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 11/15] MIPS: lantiq: Add a GPHY driver which uses the RCU syscon-mfd Hauke Mehrtens
     [not found]     ` <20170521130918.27446-12-hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org>
2017-05-23  7:52       ` Andy Shevchenko
     [not found]         ` <CAHp75VdSAGv0md8YsvwdZJX4Eo-K6Tv3TcyAVKfOmdk6De1iGQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-05-25 18:53           ` Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 12/15] MIPS: lantiq: remove old GPHY loader code Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 13/15] phy: Add an USB PHY driver for the Lantiq SoCs using the RCU module Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 14/15] MIPS: lantiq: remove old USB PHY initialisation Hauke Mehrtens
2017-05-21 13:09   ` [PATCH v2 15/15] MIPS: lantiq: Remove the arch/mips/lantiq/xway/reset.c implementation Hauke Mehrtens

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=1495445613.3558.67.camel@pengutronix.de \
    --to=p.zabel-bicnvbalz9megne8c9+irq@public.gmane.org \
    --cc=devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=hauke-5/S+JYg5SzeELgA04lAiVw@public.gmane.org \
    --cc=hauke.mehrtens-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    --cc=john-Pj+rj9U5foFAfugRpC6u6w@public.gmane.org \
    --cc=linux-mips-6z/3iImG2C8G8FEW9MqTrA@public.gmane.org \
    --cc=linux-mtd-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
    --cc=linux-spi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-watchdog-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=martin.blumenstingl-gM/Ye1E23mwN+BqQ9rBEUg@public.gmane.org \
    --cc=ralf-6z/3iImG2C8G8FEW9MqTrA@public.gmane.org \
    --cc=robh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.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).