[PATCH] staging: Add Xilinx Clocking Wizard driver

[PATCH] staging: Add Xilinx Clocking Wizard driver

[Soren Brinkmann]

Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
provides an AXI interface to dynamically reconfigure the clocking
resources of Xilinx FPGAs.

Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>
---
 drivers/staging/Kconfig                            |   2 +
 drivers/staging/Makefile                           |   1 +
 drivers/staging/clocking-wizard/Kconfig            |   9 +
 drivers/staging/clocking-wizard/Makefile           |   1 +
 drivers/staging/clocking-wizard/TODO               |  12 +
 .../clocking-wizard/clk-xlnx-clock-wizard.c        | 304 +++++++++++++++++++++
 drivers/staging/clocking-wizard/dt-binding.txt     |  30 ++
 7 files changed, 359 insertions(+)
 create mode 100644 drivers/staging/clocking-wizard/Kconfig
 create mode 100644 drivers/staging/clocking-wizard/Makefile
 create mode 100644 drivers/staging/clocking-wizard/TODO
 create mode 100644 drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
 create mode 100644 drivers/staging/clocking-wizard/dt-binding.txt

diff --git a/drivers/staging/Kconfig b/drivers/staging/Kconfig
index 35b494f5667f..7daf345bec0e 100644
--- a/drivers/staging/Kconfig
+++ b/drivers/staging/Kconfig
@@ -118,4 +118,6 @@ source "drivers/staging/skein/Kconfig"
 
 source "drivers/staging/unisys/Kconfig"
 
+source "drivers/staging/clocking-wizard/Kconfig"
+
 endif # STAGING
diff --git a/drivers/staging/Makefile b/drivers/staging/Makefile
index e66a5dbd9b02..03fec7f981cc 100644
--- a/drivers/staging/Makefile
+++ b/drivers/staging/Makefile
@@ -51,3 +51,4 @@ obj-$(CONFIG_GS_FPGABOOT)	+= gs_fpgaboot/
 obj-$(CONFIG_BT_NOKIA_H4P)	+= nokia_h4p/
 obj-$(CONFIG_CRYPTO_SKEIN)	+= skein/
 obj-$(CONFIG_UNISYSSPAR)	+= unisys/
+obj-$(CONFIG_COMMON_CLK_XLNX_CLKWZRD)	+= clocking-wizard/
diff --git a/drivers/staging/clocking-wizard/Kconfig b/drivers/staging/clocking-wizard/Kconfig
new file mode 100644
index 000000000000..3243c92f6e6d
--- /dev/null
+++ b/drivers/staging/clocking-wizard/Kconfig
@@ -0,0 +1,9 @@
+#
+# Xilinx Clocking Wizard Driver
+#
+
+config COMMON_CLK_XLNX_CLKWZRD
+	bool "Xilinx Clocking Wizard"
+	depends on COMMON_CLK && OF
+	---help---
+	  Support for the Xilinx Clocking Wizard IP core clock generator.
diff --git a/drivers/staging/clocking-wizard/Makefile b/drivers/staging/clocking-wizard/Makefile
new file mode 100644
index 000000000000..5ad352f521fe
--- /dev/null
+++ b/drivers/staging/clocking-wizard/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_COMMON_CLK_XLNX_CLKWZRD)	+= clk-xlnx-clock-wizard.o
diff --git a/drivers/staging/clocking-wizard/TODO b/drivers/staging/clocking-wizard/TODO
new file mode 100644
index 000000000000..ebe99db7d153
--- /dev/null
+++ b/drivers/staging/clocking-wizard/TODO
@@ -0,0 +1,12 @@
+TODO:
+	- support for fractional multiplier
+	- support for fractional divider (output 0 only)
+	- support for set_rate() operations (may benefit from Stephen Boyd's
+	  refactoring of the clk primitives: https://lkml.org/lkml/2014/9/5/766)
+	- review arithmetic
+	  - overflow after multiplication?
+	  - maximize accuracy before divisions
+
+Patches to:
+	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+	Sören Brinkmann <soren.brinkmann@xilinx.com>
diff --git a/drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c b/drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
new file mode 100644
index 000000000000..ad126e1c43c1
--- /dev/null
+++ b/drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
@@ -0,0 +1,304 @@
+/*
+ * Xilinx 'Clocking Wizard' driver
+ *
+ *  Copyright (C) 2013 - 2014 Xilinx
+ *
+ *  Sören Brinkmann <soren.brinkmann@xilinx.com>
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/platform_device.h>
+#include <linux/clk-provider.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/module.h>
+#include <linux/err.h>
+
+#define WZRD_NUM_OUTPUTS	7
+#define WZRD_ACLK_MAX_FREQ	250000000UL
+
+#define WZRD_CLK_CFG_REG(n)	(0x200 + 4 * n)
+
+#define WZRD_CLkOUT0_FRAC_EN	BIT(18)
+#define WZRD_CLkFBOUT_FRAC_EN	BIT(26)
+
+#define WZRD_CLKFBOUT_MULT_SHIFT	8
+#define WZRD_CLKFBOUT_MULT_MASK		(0xff << WZRD_CLKFBOUT_MULT_SHIFT)
+#define WZRD_DIVCLK_DIVIDE_SHIFT	0
+#define WZRD_DIVCLK_DIVIDE_MASK		(0xff << WZRD_DIVCLK_DIVIDE_SHIFT)
+#define WZRD_CLKOUT_DIVIDE_SHIFT	0
+#define WZRD_CLKOUT_DIVIDE_MASK		(0xff << WZRD_DIVCLK_DIVIDE_SHIFT)
+
+enum clk_wzrd_inp_clks {
+	wzrd_clk_in1,
+	wzrd_s_axi_aclk,
+	wzrd_clk_inp_max
+};
+
+enum clk_wzrd_int_clks {
+	wzrd_clk_mul,
+	wzrd_clk_mul_div,
+	wzrd_clk_int_max
+};
+
+/**
+ * struct clk_wzrd:
+ * @clk_data:		Clock data
+ * @nb:			Notifier block
+ * @base:		Memory base
+ * @clkin:		Handle to input clocks
+ * @clks_internal:	Internal clocks
+ * @clkout:		Output clocks
+ * @speed_grade:	Negated speed grade of the device
+ * @suspended:		Flag indicating power state of the device
+ */
+struct clk_wzrd {
+	struct clk_onecell_data clk_data;
+	struct notifier_block nb;
+	void __iomem *base;
+	struct clk *clkin[wzrd_clk_inp_max];
+	struct clk *clks_internal[wzrd_clk_int_max];
+	struct clk *clkout[WZRD_NUM_OUTPUTS];
+	int speed_grade;
+	bool suspended;
+};
+#define to_clk_wzrd(_nb) container_of(_nb, struct clk_wzrd, nb);
+
+/* maximum frequencies for input/output clocks per speed grade */
+static const unsigned long clk_wzrd_max_freq[] = {
+	800000000UL,
+	933000000UL,
+	1066000000UL
+};
+
+static int clk_wzrd_clk_notifier(struct notifier_block *nb, unsigned long event,
+				 void *data)
+{
+	unsigned long max;
+	struct clk_notifier_data *ndata = data;
+	struct clk_wzrd *clk_wzrd = to_clk_wzrd(nb);
+
+	if (clk_wzrd->suspended)
+		return NOTIFY_OK;
+
+	if (ndata->clk == clk_wzrd->clkin[wzrd_clk_in1])
+		max = clk_wzrd_max_freq[clk_wzrd->speed_grade - 1];
+	if (ndata->clk == clk_wzrd->clkin[wzrd_s_axi_aclk])
+		max = WZRD_ACLK_MAX_FREQ;
+
+	switch (event) {
+	case PRE_RATE_CHANGE:
+		if (ndata->new_rate > max)
+			return NOTIFY_BAD;
+		return NOTIFY_OK;
+	case POST_RATE_CHANGE:
+	case ABORT_RATE_CHANGE:
+	default:
+		return NOTIFY_DONE;
+	}
+}
+
+static int clk_wzrd_probe(struct platform_device *pdev)
+{
+	int i, ret;
+	u32 reg;
+	unsigned long rate;
+	const char *clk_name;
+	struct clk_wzrd *clk_wzrd;
+	struct resource *mem;
+	struct device_node *np = pdev->dev.of_node;
+
+	clk_wzrd = devm_kzalloc(&pdev->dev, sizeof(*clk_wzrd), GFP_KERNEL);
+	if (!clk_wzrd)
+		return -ENOMEM;
+	platform_set_drvdata(pdev, clk_wzrd);
+
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	clk_wzrd->base = devm_ioremap_resource(&pdev->dev, mem);
+	if (IS_ERR(clk_wzrd->base))
+		return PTR_ERR(clk_wzrd->base);
+
+	ret = of_property_read_u32(np, "speed-grade", &clk_wzrd->speed_grade);
+	if (!ret) {
+		clk_wzrd->speed_grade = -clk_wzrd->speed_grade;
+		if (clk_wzrd->speed_grade < 1 || clk_wzrd->speed_grade > 3) {
+			dev_warn(&pdev->dev, "invalid speed grade\n");
+			clk_wzrd->speed_grade = 0;
+		}
+	}
+
+	clk_wzrd->clkin[wzrd_clk_in1] = devm_clk_get(&pdev->dev, "clk_in1");
+	if (IS_ERR(clk_wzrd->clkin[wzrd_clk_in1])) {
+		if (clk_wzrd->clkin[wzrd_clk_in1] != ERR_PTR(-EPROBE_DEFER))
+			dev_err(&pdev->dev, "clk_in1 not found\n");
+		return PTR_ERR(clk_wzrd->clkin[wzrd_clk_in1]);
+	}
+
+	clk_wzrd->clkin[wzrd_s_axi_aclk] = devm_clk_get(&pdev->dev, "s_axi_aclk");
+	if (IS_ERR(clk_wzrd->clkin[wzrd_s_axi_aclk])) {
+		if (clk_wzrd->clkin[wzrd_s_axi_aclk] != ERR_PTR(-EPROBE_DEFER))
+			dev_err(&pdev->dev, "s_axi_aclk not found\n");
+		return PTR_ERR(clk_wzrd->clkin[wzrd_s_axi_aclk]);
+	}
+	ret = clk_prepare_enable(clk_wzrd->clkin[wzrd_s_axi_aclk]);
+	if (ret) {
+		dev_err(&pdev->dev, "enabling s_axi_aclk failed\n");
+		return ret;
+	}
+	rate = clk_get_rate(clk_wzrd->clkin[wzrd_s_axi_aclk]);
+	if (rate > WZRD_ACLK_MAX_FREQ) {
+		dev_err(&pdev->dev, "s_axi_aclk frequency too high\n");
+		ret = -EINVAL;
+		goto err_disable_clk;
+	}
+
+	/* we don't support fractional div/mul yet */
+	reg = readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) & WZRD_CLkFBOUT_FRAC_EN;
+	reg |= readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2)) & WZRD_CLkOUT0_FRAC_EN;
+	if (reg)
+		dev_warn(&pdev->dev, "fractional div/mul not supported\n");
+
+	/* register multiplier */
+	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) & WZRD_CLKFBOUT_MULT_MASK) >>
+				WZRD_CLKFBOUT_MULT_SHIFT;
+	clk_name = kasprintf(GFP_KERNEL, "%s_mul", dev_name(&pdev->dev));
+	if (!clk_name) {
+		ret = -ENOMEM;
+		goto err_disable_clk;
+	}
+	clk_wzrd->clks_internal[wzrd_clk_mul] = clk_register_fixed_factor(
+			&pdev->dev, clk_name,
+			__clk_get_name(clk_wzrd->clkin[wzrd_clk_in1]),
+			0, reg,1);
+	kfree(clk_name);
+	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul])) {
+		dev_err(&pdev->dev, "unable to register fixed-factor clock\n");
+		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul]);
+		goto err_disable_clk;
+	}
+
+	/* register div */
+	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) &
+			WZRD_DIVCLK_DIVIDE_MASK) >> WZRD_DIVCLK_DIVIDE_SHIFT;
+	clk_name = kasprintf(GFP_KERNEL, "%s_mul_div", dev_name(&pdev->dev));
+	clk_wzrd->clks_internal[wzrd_clk_mul_div] = clk_register_fixed_factor(
+			&pdev->dev, clk_name,
+			__clk_get_name(clk_wzrd->clks_internal[wzrd_clk_mul]),
+			0, 1, reg);
+	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div])) {
+		dev_err(&pdev->dev, "unable to register divider clock\n");
+		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div]);
+		goto err_rm_int_clk;
+	}
+
+	/* register div per output */
+	for (i = WZRD_NUM_OUTPUTS - 1; i >= 0 ; i--) {
+		const char *clkout_name;
+		if (of_property_read_string_index(np, "clock-output-names", i,
+					&clkout_name)) {
+			dev_err(&pdev->dev,
+					"clock output name not specified\n");
+			ret = -EINVAL;
+			goto err_rm_int_clks;
+		}
+		reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2) + i * 12) &
+				WZRD_CLKOUT_DIVIDE_MASK) >> WZRD_CLKOUT_DIVIDE_SHIFT;
+		clk_wzrd->clkout[i] = clk_register_fixed_factor(&pdev->dev,
+				clkout_name, clk_name, 0, 1, reg);
+		if (IS_ERR(clk_wzrd->clkout[i])) {
+			dev_err(&pdev->dev, "unable to register divider clock\n");
+			ret = PTR_ERR(clk_wzrd->clkout[i]);
+			goto err_rm_int_clks;
+		}
+	}
+
+	kfree(clk_name);
+
+	clk_wzrd->clk_data.clks = clk_wzrd->clkout;
+	clk_wzrd->clk_data.clk_num = ARRAY_SIZE(clk_wzrd->clkout);
+	of_clk_add_provider(np, of_clk_src_onecell_get, &clk_wzrd->clk_data);
+
+	if (clk_wzrd->speed_grade) {
+		clk_wzrd->nb.notifier_call = clk_wzrd_clk_notifier;
+
+		ret = clk_notifier_register(clk_wzrd->clkin[wzrd_clk_in1],
+					    &clk_wzrd->nb);
+		if (ret)
+			dev_warn(&pdev->dev,
+					"unable to register clock notifier\n");
+
+		ret = clk_notifier_register(clk_wzrd->clkin[wzrd_s_axi_aclk],
+					    &clk_wzrd->nb);
+		if (ret)
+			dev_warn(&pdev->dev,
+					"unable to register clock notifier\n");
+	}
+
+	return 0;
+
+err_rm_int_clks:
+	clk_unregister(clk_wzrd->clks_internal[1]);
+err_rm_int_clk:
+	kfree(clk_name);
+	clk_unregister(clk_wzrd->clks_internal[0]);
+err_disable_clk:
+	clk_disable_unprepare(clk_wzrd->clkin[wzrd_s_axi_aclk]);
+
+	return ret;
+}
+
+static int clk_wzrd_remove(struct platform_device *pdev)
+{
+	int i;
+	struct clk_wzrd *clk_wzrd = platform_get_drvdata(pdev);
+
+	of_clk_del_provider(pdev->dev.of_node);
+
+	for (i = 0; i < WZRD_NUM_OUTPUTS; i++)
+		clk_unregister(clk_wzrd->clkout[i]);
+	for (i = 0; i < wzrd_clk_int_max; i++)
+		clk_unregister(clk_wzrd->clks_internal[i]);
+
+	if (clk_wzrd->speed_grade) {
+		clk_notifier_unregister(clk_wzrd->clkin[wzrd_s_axi_aclk],
+					&clk_wzrd->nb);
+		clk_notifier_unregister(clk_wzrd->clkin[wzrd_clk_in1],
+					&clk_wzrd->nb);
+	}
+
+	clk_disable_unprepare(clk_wzrd->clkin[wzrd_s_axi_aclk]);
+
+	return 0;
+}
+
+static const struct of_device_id clk_wzrd_ids[] = {
+	{ .compatible = "xlnx,clocking-wizard" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, clk_wzrd_ids);
+
+static struct platform_driver clk_wzrd_driver = {
+	.driver = {
+		.name = "clk-wizard",
+		.of_match_table = clk_wzrd_ids,
+	},
+	.probe = clk_wzrd_probe,
+	.remove = clk_wzrd_remove,
+};
+module_platform_driver(clk_wzrd_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Soeren Brinkmann <soren.brinkmann@xilinx.com");
+MODULE_DESCRIPTION("Driver for the Xilinx Clocking Wizard IP core");
diff --git a/drivers/staging/clocking-wizard/dt-binding.txt b/drivers/staging/clocking-wizard/dt-binding.txt
new file mode 100644
index 000000000000..cc3cc5e91440
--- /dev/null
+++ b/drivers/staging/clocking-wizard/dt-binding.txt
@@ -0,0 +1,30 @@
+Binding for Xilinx Clocking Wizard IP Core
+
+This binding uses the common clock binding[1]. Details about the devices can be
+found in the product guide[2].
+
+[1] Documentation/devicetree/bindings/clock/clock-bindings.txt
+[2] Clocking Wizard Product Guide
+http://www.xilinx.com/support/documentation/ip_documentation/clk_wiz/v5_1/pg065-clk-wiz.pdf
+
+Required properties:
+ - compatible: Must be 'xlnx,clocking-wizard'
+ - reg: Base and size of the cores register space
+ - clocks: Handle to input clock
+ - clock-names: Tuple containing 'clk_in1' and 's_axi_aclk'
+ - clock-output-names: Names for the output clocks
+
+Optional properties:
+ - speed-grade: Speed grade of the device
+
+Example:
+	clock-generator@40040000 {
+		reg = <0x40040000 0x1000>;
+		compatible = "xlnx,clocking-wizard";
+		speed-grade = <(-1)>;
+		clock-names = "clk_in1", "s_axi_aclk";
+		clocks = <&clkc 15>, <&clkc 15>;
+		clock-output-names = "clk_out0", "clk_out1", "clk_out2",
+				     "clk_out3", "clk_out4", "clk_out5",
+				     "clk_out6", "clk_out7";
+	};
-- 
2.1.1

Re: [PATCH] staging: Add Xilinx Clocking Wizard driver

[Greg Kroah-Hartman]

On Wed, Oct 01, 2014 at 10:21:48AM -0700, Soren Brinkmann wrote:
> Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
> provides an AXI interface to dynamically reconfigure the clocking
> resources of Xilinx FPGAs.

Why not just do the few things you have on the TODO list and get it
merged to the "proper" part of the kernel, keeping it out of the staging
tree?

thanks,

greg k-h

Re: [PATCH] staging: Add Xilinx Clocking Wizard driver

[Sören Brinkmann]

On Wed, 2014-10-01 at 10:39AM -0700, Greg Kroah-Hartman wrote:
> On Wed, Oct 01, 2014 at 10:21:48AM -0700, Soren Brinkmann wrote:
> > Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
> > provides an AXI interface to dynamically reconfigure the clocking
> > resources of Xilinx FPGAs.
> 
> Why not just do the few things you have on the TODO list and get it
> merged to the "proper" part of the kernel, keeping it out of the staging
> tree?

The few things are not that easy and as I mention in the TODO file,
there are some patches on LKML that would greatly simplify this
driver/reduce the need for code duplication.
I thought this is a good way to wait for those parts to mature while
people could already use the driver as passive part in their clock tree.

	Sören

Re: [PATCH] staging: Add Xilinx Clocking Wizard driver

[Greg Kroah-Hartman]

On Wed, Oct 01, 2014 at 10:46:16AM -0700, Sören Brinkmann wrote:
> On Wed, 2014-10-01 at 10:39AM -0700, Greg Kroah-Hartman wrote:
> > On Wed, Oct 01, 2014 at 10:21:48AM -0700, Soren Brinkmann wrote:
> > > Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
> > > provides an AXI interface to dynamically reconfigure the clocking
> > > resources of Xilinx FPGAs.
> > 
> > Why not just do the few things you have on the TODO list and get it
> > merged to the "proper" part of the kernel, keeping it out of the staging
> > tree?
> 
> The few things are not that easy and as I mention in the TODO file,
> there are some patches on LKML that would greatly simplify this
> driver/reduce the need for code duplication.
> I thought this is a good way to wait for those parts to mature while
> people could already use the driver as passive part in their clock tree.

Ok, fair enough, as long as you keep working on the driver to get it out
of staging, I don't have an objection to it.  I'll go queue it up later
today.

greg k-h

Re: [PATCH] staging: Add Xilinx Clocking Wizard driver

[Sören Brinkmann]

On Wed, 2014-10-01 at 10:57AM -0700, Greg Kroah-Hartman wrote:
> On Wed, Oct 01, 2014 at 10:46:16AM -0700, Sören Brinkmann wrote:
> > On Wed, 2014-10-01 at 10:39AM -0700, Greg Kroah-Hartman wrote:
> > > On Wed, Oct 01, 2014 at 10:21:48AM -0700, Soren Brinkmann wrote:
> > > > Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
> > > > provides an AXI interface to dynamically reconfigure the clocking
> > > > resources of Xilinx FPGAs.
> > > 
> > > Why not just do the few things you have on the TODO list and get it
> > > merged to the "proper" part of the kernel, keeping it out of the staging
> > > tree?
> > 
> > The few things are not that easy and as I mention in the TODO file,
> > there are some patches on LKML that would greatly simplify this
> > driver/reduce the need for code duplication.
> > I thought this is a good way to wait for those parts to mature while
> > people could already use the driver as passive part in their clock tree.
> 
> Ok, fair enough, as long as you keep working on the driver to get it out
> of staging, I don't have an objection to it.  I'll go queue it up later
> today.

Thank you.

	Sören

Re: [PATCH] staging: Add Xilinx Clocking Wizard driver

[Dan Carpenter]

On Wed, Oct 01, 2014 at 10:21:48AM -0700, Soren Brinkmann wrote:
> +enum clk_wzrd_inp_clks {
> +	wzrd_clk_in1,
> +	wzrd_s_axi_aclk,
> +	wzrd_clk_inp_max
> +};
> +
> +enum clk_wzrd_int_clks {
> +	wzrd_clk_mul,
> +	wzrd_clk_mul_div,
> +	wzrd_clk_int_max
> +};
> +
> +/**
> + * struct clk_wzrd:
> + * @clk_data:		Clock data
> + * @nb:			Notifier block
> + * @base:		Memory base
> + * @clkin:		Handle to input clocks
> + * @clks_internal:	Internal clocks
> + * @clkout:		Output clocks
> + * @speed_grade:	Negated speed grade of the device
> + * @suspended:		Flag indicating power state of the device
> + */
> +struct clk_wzrd {
> +	struct clk_onecell_data clk_data;
> +	struct notifier_block nb;
> +	void __iomem *base;
> +	struct clk *clkin[wzrd_clk_inp_max];
> +	struct clk *clks_internal[wzrd_clk_int_max];

There is no advantage to using these arrays here.  It just makes the
code more complicated to look at:

before:		clk_wzrd->clks_internal[wzrd_clk_mul_div] = ...
 after:		clk_wzrd->mul_div = ...

> +	struct clk *clkout[WZRD_NUM_OUTPUTS];

This array makes sense, though.

> +	int speed_grade;
> +	bool suspended;

suspended is always zero.  Delete it.

> +};
> +#define to_clk_wzrd(_nb) container_of(_nb, struct clk_wzrd, nb);
> +
> +/* maximum frequencies for input/output clocks per speed grade */
> +static const unsigned long clk_wzrd_max_freq[] = {
> +	800000000UL,
> +	933000000UL,
> +	1066000000UL
> +};
> +
> +static int clk_wzrd_clk_notifier(struct notifier_block *nb, unsigned long event,
> +				 void *data)
> +{
> +	unsigned long max;
> +	struct clk_notifier_data *ndata = data;
> +	struct clk_wzrd *clk_wzrd = to_clk_wzrd(nb);
> +
> +	if (clk_wzrd->suspended)
> +		return NOTIFY_OK;
> +
> +	if (ndata->clk == clk_wzrd->clkin[wzrd_clk_in1])
> +		max = clk_wzrd_max_freq[clk_wzrd->speed_grade - 1];
> +	if (ndata->clk == clk_wzrd->clkin[wzrd_s_axi_aclk])
> +		max = WZRD_ACLK_MAX_FREQ;
> +
> +	switch (event) {
> +	case PRE_RATE_CHANGE:
> +		if (ndata->new_rate > max)
> +			return NOTIFY_BAD;
> +		return NOTIFY_OK;
> +	case POST_RATE_CHANGE:
> +	case ABORT_RATE_CHANGE:
> +	default:
> +		return NOTIFY_DONE;
> +	}
> +}
> +
> +static int clk_wzrd_probe(struct platform_device *pdev)
> +{
> +	int i, ret;
> +	u32 reg;
> +	unsigned long rate;
> +	const char *clk_name;
> +	struct clk_wzrd *clk_wzrd;
> +	struct resource *mem;
> +	struct device_node *np = pdev->dev.of_node;
> +
> +	clk_wzrd = devm_kzalloc(&pdev->dev, sizeof(*clk_wzrd), GFP_KERNEL);
> +	if (!clk_wzrd)
> +		return -ENOMEM;
> +	platform_set_drvdata(pdev, clk_wzrd);
> +
> +	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> +	clk_wzrd->base = devm_ioremap_resource(&pdev->dev, mem);
> +	if (IS_ERR(clk_wzrd->base))
> +		return PTR_ERR(clk_wzrd->base);
> +
> +	ret = of_property_read_u32(np, "speed-grade", &clk_wzrd->speed_grade);
> +	if (!ret) {
> +		clk_wzrd->speed_grade = -clk_wzrd->speed_grade;
> +		if (clk_wzrd->speed_grade < 1 || clk_wzrd->speed_grade > 3) {
> +			dev_warn(&pdev->dev, "invalid speed grade\n");

Print what it was.

> +			clk_wzrd->speed_grade = 0;
> +		}
> +	}
> +
> +	clk_wzrd->clkin[wzrd_clk_in1] = devm_clk_get(&pdev->dev, "clk_in1");
> +	if (IS_ERR(clk_wzrd->clkin[wzrd_clk_in1])) {
> +		if (clk_wzrd->clkin[wzrd_clk_in1] != ERR_PTR(-EPROBE_DEFER))
> +			dev_err(&pdev->dev, "clk_in1 not found\n");
> +		return PTR_ERR(clk_wzrd->clkin[wzrd_clk_in1]);
> +	}
> +
> +	clk_wzrd->clkin[wzrd_s_axi_aclk] = devm_clk_get(&pdev->dev, "s_axi_aclk");
> +	if (IS_ERR(clk_wzrd->clkin[wzrd_s_axi_aclk])) {
> +		if (clk_wzrd->clkin[wzrd_s_axi_aclk] != ERR_PTR(-EPROBE_DEFER))
> +			dev_err(&pdev->dev, "s_axi_aclk not found\n");
> +		return PTR_ERR(clk_wzrd->clkin[wzrd_s_axi_aclk]);
> +	}
> +	ret = clk_prepare_enable(clk_wzrd->clkin[wzrd_s_axi_aclk]);
> +	if (ret) {
> +		dev_err(&pdev->dev, "enabling s_axi_aclk failed\n");
> +		return ret;
> +	}
> +	rate = clk_get_rate(clk_wzrd->clkin[wzrd_s_axi_aclk]);
> +	if (rate > WZRD_ACLK_MAX_FREQ) {
> +		dev_err(&pdev->dev, "s_axi_aclk frequency too high\n");

Print what it was.

> +		ret = -EINVAL;
> +		goto err_disable_clk;
> +	}
> +
> +	/* we don't support fractional div/mul yet */
> +	reg = readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) & WZRD_CLkFBOUT_FRAC_EN;
> +	reg |= readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2)) & WZRD_CLkOUT0_FRAC_EN;
> +	if (reg)
> +		dev_warn(&pdev->dev, "fractional div/mul not supported\n");
> +
> +	/* register multiplier */
> +	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) & WZRD_CLKFBOUT_MULT_MASK) >>
> +				WZRD_CLKFBOUT_MULT_SHIFT;
> +	clk_name = kasprintf(GFP_KERNEL, "%s_mul", dev_name(&pdev->dev));
> +	if (!clk_name) {
> +		ret = -ENOMEM;
> +		goto err_disable_clk;
> +	}
> +	clk_wzrd->clks_internal[wzrd_clk_mul] = clk_register_fixed_factor(
> +			&pdev->dev, clk_name,
> +			__clk_get_name(clk_wzrd->clkin[wzrd_clk_in1]),
> +			0, reg,1);
> +	kfree(clk_name);
> +	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul])) {
> +		dev_err(&pdev->dev, "unable to register fixed-factor clock\n");
> +		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul]);
> +		goto err_disable_clk;
> +	}
> +
> +	/* register div */
> +	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) &
> +			WZRD_DIVCLK_DIVIDE_MASK) >> WZRD_DIVCLK_DIVIDE_SHIFT;
> +	clk_name = kasprintf(GFP_KERNEL, "%s_mul_div", dev_name(&pdev->dev));
> +	clk_wzrd->clks_internal[wzrd_clk_mul_div] = clk_register_fixed_factor(
> +			&pdev->dev, clk_name,
> +			__clk_get_name(clk_wzrd->clks_internal[wzrd_clk_mul]),
> +			0, 1, reg);
> +	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div])) {
> +		dev_err(&pdev->dev, "unable to register divider clock\n");
> +		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div]);
> +		goto err_rm_int_clk;
> +	}
> +
> +	/* register div per output */
> +	for (i = WZRD_NUM_OUTPUTS - 1; i >= 0 ; i--) {
> +		const char *clkout_name;
> +		if (of_property_read_string_index(np, "clock-output-names", i,
> +					&clkout_name)) {
> +			dev_err(&pdev->dev,
> +					"clock output name not specified\n");

Run checkpatch.pl --strict over this code.

> +			ret = -EINVAL;
> +			goto err_rm_int_clks;
> +		}
> +		reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2) + i * 12) &
> +				WZRD_CLKOUT_DIVIDE_MASK) >> WZRD_CLKOUT_DIVIDE_SHIFT;
> +		clk_wzrd->clkout[i] = clk_register_fixed_factor(&pdev->dev,
> +				clkout_name, clk_name, 0, 1, reg);

Alignment is hard to look at.

> +		if (IS_ERR(clk_wzrd->clkout[i])) {
> +			dev_err(&pdev->dev, "unable to register divider clock\n");
> +			ret = PTR_ERR(clk_wzrd->clkout[i]);
> +			goto err_rm_int_clks;
> +		}

The error handling for this loop should unregister ->clkout[i + 1] etc.
Why is does this loop count backwards, just out of curiosity?

regards,
dan carpenter

Re: [PATCH] staging: Add Xilinx Clocking Wizard driver

[Sören Brinkmann]

On Wed, 2014-10-01 at 09:58PM +0300, Dan Carpenter wrote:
> On Wed, Oct 01, 2014 at 10:21:48AM -0700, Soren Brinkmann wrote:
> > +enum clk_wzrd_inp_clks {
> > +	wzrd_clk_in1,
> > +	wzrd_s_axi_aclk,
> > +	wzrd_clk_inp_max
> > +};
> > +
> > +enum clk_wzrd_int_clks {
> > +	wzrd_clk_mul,
> > +	wzrd_clk_mul_div,
> > +	wzrd_clk_int_max
> > +};
> > +
> > +/**
> > + * struct clk_wzrd:
> > + * @clk_data:		Clock data
> > + * @nb:			Notifier block
> > + * @base:		Memory base
> > + * @clkin:		Handle to input clocks
> > + * @clks_internal:	Internal clocks
> > + * @clkout:		Output clocks
> > + * @speed_grade:	Negated speed grade of the device
> > + * @suspended:		Flag indicating power state of the device
> > + */
> > +struct clk_wzrd {
> > +	struct clk_onecell_data clk_data;
> > +	struct notifier_block nb;
> > +	void __iomem *base;
> > +	struct clk *clkin[wzrd_clk_inp_max];
> > +	struct clk *clks_internal[wzrd_clk_int_max];
> 
> There is no advantage to using these arrays here.  It just makes the
> code more complicated to look at:
> 
> before:		clk_wzrd->clks_internal[wzrd_clk_mul_div] = ...
>  after:		clk_wzrd->mul_div = ...

Yeah, I was thinking about it, guess you're right.

> 
> > +	struct clk *clkout[WZRD_NUM_OUTPUTS];
> 
> This array makes sense, though.
> 
> > +	int speed_grade;
> > +	bool suspended;
> 
> suspended is always zero.  Delete it.

Right, because I forgot to implement the dev_pm_ops. It'll be fixed that
way.

[...]
> > +static int clk_wzrd_probe(struct platform_device *pdev)
> > +{
> > +	int i, ret;
> > +	u32 reg;
> > +	unsigned long rate;
> > +	const char *clk_name;
> > +	struct clk_wzrd *clk_wzrd;
> > +	struct resource *mem;
> > +	struct device_node *np = pdev->dev.of_node;
> > +
> > +	clk_wzrd = devm_kzalloc(&pdev->dev, sizeof(*clk_wzrd), GFP_KERNEL);
> > +	if (!clk_wzrd)
> > +		return -ENOMEM;
> > +	platform_set_drvdata(pdev, clk_wzrd);
> > +
> > +	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> > +	clk_wzrd->base = devm_ioremap_resource(&pdev->dev, mem);
> > +	if (IS_ERR(clk_wzrd->base))
> > +		return PTR_ERR(clk_wzrd->base);
> > +
> > +	ret = of_property_read_u32(np, "speed-grade", &clk_wzrd->speed_grade);
> > +	if (!ret) {
> > +		clk_wzrd->speed_grade = -clk_wzrd->speed_grade;
> > +		if (clk_wzrd->speed_grade < 1 || clk_wzrd->speed_grade > 3) {
> > +			dev_warn(&pdev->dev, "invalid speed grade\n");
> 
> Print what it was.
> 
> > +			clk_wzrd->speed_grade = 0;
> > +		}
> > +	}
> > +
> > +	clk_wzrd->clkin[wzrd_clk_in1] = devm_clk_get(&pdev->dev, "clk_in1");
> > +	if (IS_ERR(clk_wzrd->clkin[wzrd_clk_in1])) {
> > +		if (clk_wzrd->clkin[wzrd_clk_in1] != ERR_PTR(-EPROBE_DEFER))
> > +			dev_err(&pdev->dev, "clk_in1 not found\n");
> > +		return PTR_ERR(clk_wzrd->clkin[wzrd_clk_in1]);
> > +	}
> > +
> > +	clk_wzrd->clkin[wzrd_s_axi_aclk] = devm_clk_get(&pdev->dev, "s_axi_aclk");
> > +	if (IS_ERR(clk_wzrd->clkin[wzrd_s_axi_aclk])) {
> > +		if (clk_wzrd->clkin[wzrd_s_axi_aclk] != ERR_PTR(-EPROBE_DEFER))
> > +			dev_err(&pdev->dev, "s_axi_aclk not found\n");
> > +		return PTR_ERR(clk_wzrd->clkin[wzrd_s_axi_aclk]);
> > +	}
> > +	ret = clk_prepare_enable(clk_wzrd->clkin[wzrd_s_axi_aclk]);
> > +	if (ret) {
> > +		dev_err(&pdev->dev, "enabling s_axi_aclk failed\n");
> > +		return ret;
> > +	}
> > +	rate = clk_get_rate(clk_wzrd->clkin[wzrd_s_axi_aclk]);
> > +	if (rate > WZRD_ACLK_MAX_FREQ) {
> > +		dev_err(&pdev->dev, "s_axi_aclk frequency too high\n");
> 
> Print what it was.
> 
> > +		ret = -EINVAL;
> > +		goto err_disable_clk;
> > +	}
> > +
> > +	/* we don't support fractional div/mul yet */
> > +	reg = readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) & WZRD_CLkFBOUT_FRAC_EN;
> > +	reg |= readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2)) & WZRD_CLkOUT0_FRAC_EN;
> > +	if (reg)
> > +		dev_warn(&pdev->dev, "fractional div/mul not supported\n");
> > +
> > +	/* register multiplier */
> > +	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) & WZRD_CLKFBOUT_MULT_MASK) >>
> > +				WZRD_CLKFBOUT_MULT_SHIFT;
> > +	clk_name = kasprintf(GFP_KERNEL, "%s_mul", dev_name(&pdev->dev));
> > +	if (!clk_name) {
> > +		ret = -ENOMEM;
> > +		goto err_disable_clk;
> > +	}
> > +	clk_wzrd->clks_internal[wzrd_clk_mul] = clk_register_fixed_factor(
> > +			&pdev->dev, clk_name,
> > +			__clk_get_name(clk_wzrd->clkin[wzrd_clk_in1]),
> > +			0, reg,1);
> > +	kfree(clk_name);
> > +	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul])) {
> > +		dev_err(&pdev->dev, "unable to register fixed-factor clock\n");
> > +		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul]);
> > +		goto err_disable_clk;
> > +	}
> > +
> > +	/* register div */
> > +	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) &
> > +			WZRD_DIVCLK_DIVIDE_MASK) >> WZRD_DIVCLK_DIVIDE_SHIFT;
> > +	clk_name = kasprintf(GFP_KERNEL, "%s_mul_div", dev_name(&pdev->dev));
> > +	clk_wzrd->clks_internal[wzrd_clk_mul_div] = clk_register_fixed_factor(
> > +			&pdev->dev, clk_name,
> > +			__clk_get_name(clk_wzrd->clks_internal[wzrd_clk_mul]),
> > +			0, 1, reg);
> > +	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div])) {
> > +		dev_err(&pdev->dev, "unable to register divider clock\n");
> > +		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div]);
> > +		goto err_rm_int_clk;
> > +	}
> > +
> > +	/* register div per output */
> > +	for (i = WZRD_NUM_OUTPUTS - 1; i >= 0 ; i--) {
> > +		const char *clkout_name;
> > +		if (of_property_read_string_index(np, "clock-output-names", i,
> > +					&clkout_name)) {
> > +			dev_err(&pdev->dev,
> > +					"clock output name not specified\n");
> 
> Run checkpatch.pl --strict over this code.
> 
> > +			ret = -EINVAL;
> > +			goto err_rm_int_clks;
> > +		}
> > +		reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2) + i * 12) &
> > +				WZRD_CLKOUT_DIVIDE_MASK) >> WZRD_CLKOUT_DIVIDE_SHIFT;
> > +		clk_wzrd->clkout[i] = clk_register_fixed_factor(&pdev->dev,
> > +				clkout_name, clk_name, 0, 1, reg);
> 
> Alignment is hard to look at.
> 
> > +		if (IS_ERR(clk_wzrd->clkout[i])) {
> > +			dev_err(&pdev->dev, "unable to register divider clock\n");
> > +			ret = PTR_ERR(clk_wzrd->clkout[i]);
> > +			goto err_rm_int_clks;
> > +		}
> 
> The error handling for this loop should unregister ->clkout[i + 1] etc.
> Why is does this loop count backwards, just out of curiosity?

Yeah, lazy error handling :) I thought, if reading
clock-output-names[MAX] succeeds I'd could save checking some stuff
since it would guarantee that the other output-names existed too.
Unfortunately clk_register() can still fail.

	Sören

[PATCH v2] staging: Add Xilinx Clocking Wizard driver

[Soren Brinkmann]

Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
provides an AXI interface to dynamically reconfigure the clocking
resources of Xilinx FPGAs.

Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>
---
Hi Greg, Dan,

I fixed the things Dan pointed out, please take this v2 instead of the original
patch.

Also, don't get caught up too much with style issues around the code that
registers the clocks. To support set_rate() opeartions this will have to change
anyway. It doesn't make a lot of sense to spend much time on those parts.

	Thanks,
	Sören

v2:
 - implement dev_pm_ops
 - don't use array for clock inputs
 - add more information in error output
 - fix style issues
 - fix error path
---
 drivers/staging/Kconfig                            |   2 +
 drivers/staging/Makefile                           |   1 +
 drivers/staging/clocking-wizard/Kconfig            |   9 +
 drivers/staging/clocking-wizard/Makefile           |   1 +
 drivers/staging/clocking-wizard/TODO               |  12 +
 .../clocking-wizard/clk-xlnx-clock-wizard.c        | 336 +++++++++++++++++++++
 drivers/staging/clocking-wizard/dt-binding.txt     |  30 ++
 7 files changed, 391 insertions(+)
 create mode 100644 drivers/staging/clocking-wizard/Kconfig
 create mode 100644 drivers/staging/clocking-wizard/Makefile
 create mode 100644 drivers/staging/clocking-wizard/TODO
 create mode 100644 drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
 create mode 100644 drivers/staging/clocking-wizard/dt-binding.txt

diff --git a/drivers/staging/Kconfig b/drivers/staging/Kconfig
index 35b494f5667f..7daf345bec0e 100644
--- a/drivers/staging/Kconfig
+++ b/drivers/staging/Kconfig
@@ -118,4 +118,6 @@ source "drivers/staging/skein/Kconfig"
 
 source "drivers/staging/unisys/Kconfig"
 
+source "drivers/staging/clocking-wizard/Kconfig"
+
 endif # STAGING
diff --git a/drivers/staging/Makefile b/drivers/staging/Makefile
index e66a5dbd9b02..03fec7f981cc 100644
--- a/drivers/staging/Makefile
+++ b/drivers/staging/Makefile
@@ -51,3 +51,4 @@ obj-$(CONFIG_GS_FPGABOOT)	+= gs_fpgaboot/
 obj-$(CONFIG_BT_NOKIA_H4P)	+= nokia_h4p/
 obj-$(CONFIG_CRYPTO_SKEIN)	+= skein/
 obj-$(CONFIG_UNISYSSPAR)	+= unisys/
+obj-$(CONFIG_COMMON_CLK_XLNX_CLKWZRD)	+= clocking-wizard/
diff --git a/drivers/staging/clocking-wizard/Kconfig b/drivers/staging/clocking-wizard/Kconfig
new file mode 100644
index 000000000000..3243c92f6e6d
--- /dev/null
+++ b/drivers/staging/clocking-wizard/Kconfig
@@ -0,0 +1,9 @@
+#
+# Xilinx Clocking Wizard Driver
+#
+
+config COMMON_CLK_XLNX_CLKWZRD
+	bool "Xilinx Clocking Wizard"
+	depends on COMMON_CLK && OF
+	---help---
+	  Support for the Xilinx Clocking Wizard IP core clock generator.
diff --git a/drivers/staging/clocking-wizard/Makefile b/drivers/staging/clocking-wizard/Makefile
new file mode 100644
index 000000000000..5ad352f521fe
--- /dev/null
+++ b/drivers/staging/clocking-wizard/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_COMMON_CLK_XLNX_CLKWZRD)	+= clk-xlnx-clock-wizard.o
diff --git a/drivers/staging/clocking-wizard/TODO b/drivers/staging/clocking-wizard/TODO
new file mode 100644
index 000000000000..ebe99db7d153
--- /dev/null
+++ b/drivers/staging/clocking-wizard/TODO
@@ -0,0 +1,12 @@
+TODO:
+	- support for fractional multiplier
+	- support for fractional divider (output 0 only)
+	- support for set_rate() operations (may benefit from Stephen Boyd's
+	  refactoring of the clk primitives: https://lkml.org/lkml/2014/9/5/766)
+	- review arithmetic
+	  - overflow after multiplication?
+	  - maximize accuracy before divisions
+
+Patches to:
+	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+	Sören Brinkmann <soren.brinkmann@xilinx.com>
diff --git a/drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c b/drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
new file mode 100644
index 000000000000..d807058b6f9f
--- /dev/null
+++ b/drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
@@ -0,0 +1,336 @@
+/*
+ * Xilinx 'Clocking Wizard' driver
+ *
+ *  Copyright (C) 2013 - 2014 Xilinx
+ *
+ *  Sören Brinkmann <soren.brinkmann@xilinx.com>
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/platform_device.h>
+#include <linux/clk-provider.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/module.h>
+#include <linux/err.h>
+
+#define WZRD_NUM_OUTPUTS	7
+#define WZRD_ACLK_MAX_FREQ	250000000UL
+
+#define WZRD_CLK_CFG_REG(n)	(0x200 + 4 * n)
+
+#define WZRD_CLkOUT0_FRAC_EN	BIT(18)
+#define WZRD_CLkFBOUT_FRAC_EN	BIT(26)
+
+#define WZRD_CLKFBOUT_MULT_SHIFT	8
+#define WZRD_CLKFBOUT_MULT_MASK		(0xff << WZRD_CLKFBOUT_MULT_SHIFT)
+#define WZRD_DIVCLK_DIVIDE_SHIFT	0
+#define WZRD_DIVCLK_DIVIDE_MASK		(0xff << WZRD_DIVCLK_DIVIDE_SHIFT)
+#define WZRD_CLKOUT_DIVIDE_SHIFT	0
+#define WZRD_CLKOUT_DIVIDE_MASK		(0xff << WZRD_DIVCLK_DIVIDE_SHIFT)
+
+enum clk_wzrd_int_clks {
+	wzrd_clk_mul,
+	wzrd_clk_mul_div,
+	wzrd_clk_int_max
+};
+
+/**
+ * struct clk_wzrd:
+ * @clk_data:		Clock data
+ * @nb:			Notifier block
+ * @base:		Memory base
+ * @clk_in1:		Handle to input clock 'clk_in1'
+ * @axi_clk:		Handle to input clock 's_axi_aclk'
+ * @clks_internal:	Internal clocks
+ * @clkout:		Output clocks
+ * @speed_grade:	Negated speed grade of the device
+ * @suspended:		Flag indicating power state of the device
+ */
+struct clk_wzrd {
+	struct clk_onecell_data clk_data;
+	struct notifier_block nb;
+	void __iomem *base;
+	struct clk *clk_in1;
+	struct clk *axi_clk;
+	struct clk *clks_internal[wzrd_clk_int_max];
+	struct clk *clkout[WZRD_NUM_OUTPUTS];
+	int speed_grade;
+	bool suspended;
+};
+#define to_clk_wzrd(_nb) container_of(_nb, struct clk_wzrd, nb)
+
+/* maximum frequencies for input/output clocks per speed grade */
+static const unsigned long clk_wzrd_max_freq[] = {
+	800000000UL,
+	933000000UL,
+	1066000000UL
+};
+
+static int clk_wzrd_clk_notifier(struct notifier_block *nb, unsigned long event,
+				 void *data)
+{
+	unsigned long max;
+	struct clk_notifier_data *ndata = data;
+	struct clk_wzrd *clk_wzrd = to_clk_wzrd(nb);
+
+	if (clk_wzrd->suspended)
+		return NOTIFY_OK;
+
+	if (ndata->clk == clk_wzrd->clk_in1)
+		max = clk_wzrd_max_freq[clk_wzrd->speed_grade - 1];
+	if (ndata->clk == clk_wzrd->axi_clk)
+		max = WZRD_ACLK_MAX_FREQ;
+
+	switch (event) {
+	case PRE_RATE_CHANGE:
+		if (ndata->new_rate > max)
+			return NOTIFY_BAD;
+		return NOTIFY_OK;
+	case POST_RATE_CHANGE:
+	case ABORT_RATE_CHANGE:
+	default:
+		return NOTIFY_DONE;
+	}
+}
+
+static int __maybe_unused clk_wzrd_suspend(struct device *dev)
+{
+	struct clk_wzrd *clk_wzrd = dev_get_drvdata(dev);
+
+	clk_disable_unprepare(clk_wzrd->axi_clk);
+	clk_wzrd->suspended = true;
+
+	return 0;
+}
+
+static int __maybe_unused clk_wzrd_resume(struct device *dev)
+{
+	int ret;
+	struct clk_wzrd *clk_wzrd = dev_get_drvdata(dev);
+
+	ret = clk_prepare_enable(clk_wzrd->axi_clk);
+	if (ret) {
+		dev_err(dev, "unable to enable s_axi_aclk\n");
+		return ret;
+	}
+
+	clk_wzrd->suspended = false;
+
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(clk_wzrd_dev_pm_ops, clk_wzrd_suspend,
+			 clk_wzrd_resume);
+
+static int clk_wzrd_probe(struct platform_device *pdev)
+{
+	int i, ret;
+	u32 reg;
+	unsigned long rate;
+	const char *clk_name;
+	struct clk_wzrd *clk_wzrd;
+	struct resource *mem;
+	struct device_node *np = pdev->dev.of_node;
+
+	clk_wzrd = devm_kzalloc(&pdev->dev, sizeof(*clk_wzrd), GFP_KERNEL);
+	if (!clk_wzrd)
+		return -ENOMEM;
+	platform_set_drvdata(pdev, clk_wzrd);
+
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	clk_wzrd->base = devm_ioremap_resource(&pdev->dev, mem);
+	if (IS_ERR(clk_wzrd->base))
+		return PTR_ERR(clk_wzrd->base);
+
+	ret = of_property_read_u32(np, "speed-grade", &clk_wzrd->speed_grade);
+	if (!ret) {
+		clk_wzrd->speed_grade = -clk_wzrd->speed_grade;
+		if (clk_wzrd->speed_grade < 1 || clk_wzrd->speed_grade > 3) {
+			dev_warn(&pdev->dev, "invalid speed grade '%d'\n",
+				 -clk_wzrd->speed_grade);
+			clk_wzrd->speed_grade = 0;
+		}
+	}
+
+	clk_wzrd->clk_in1 = devm_clk_get(&pdev->dev, "clk_in1");
+	if (IS_ERR(clk_wzrd->clk_in1)) {
+		if (clk_wzrd->clk_in1 != ERR_PTR(-EPROBE_DEFER))
+			dev_err(&pdev->dev, "clk_in1 not found\n");
+		return PTR_ERR(clk_wzrd->clk_in1);
+	}
+
+	clk_wzrd->axi_clk = devm_clk_get(&pdev->dev, "s_axi_aclk");
+	if (IS_ERR(clk_wzrd->axi_clk)) {
+		if (clk_wzrd->axi_clk != ERR_PTR(-EPROBE_DEFER))
+			dev_err(&pdev->dev, "s_axi_aclk not found\n");
+		return PTR_ERR(clk_wzrd->axi_clk);
+	}
+	ret = clk_prepare_enable(clk_wzrd->axi_clk);
+	if (ret) {
+		dev_err(&pdev->dev, "enabling s_axi_aclk failed\n");
+		return ret;
+	}
+	rate = clk_get_rate(clk_wzrd->axi_clk);
+	if (rate > WZRD_ACLK_MAX_FREQ) {
+		dev_err(&pdev->dev, "s_axi_aclk frequency (%lu) too high\n",
+			rate);
+		ret = -EINVAL;
+		goto err_disable_clk;
+	}
+
+	/* we don't support fractional div/mul yet */
+	reg = readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) &
+		    WZRD_CLkFBOUT_FRAC_EN;
+	reg |= readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2)) &
+		     WZRD_CLkOUT0_FRAC_EN;
+	if (reg)
+		dev_warn(&pdev->dev, "fractional div/mul not supported\n");
+
+	/* register multiplier */
+	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) &
+		     WZRD_CLKFBOUT_MULT_MASK) >> WZRD_CLKFBOUT_MULT_SHIFT;
+	clk_name = kasprintf(GFP_KERNEL, "%s_mul", dev_name(&pdev->dev));
+	if (!clk_name) {
+		ret = -ENOMEM;
+		goto err_disable_clk;
+	}
+	clk_wzrd->clks_internal[wzrd_clk_mul] = clk_register_fixed_factor(
+			&pdev->dev, clk_name,
+			__clk_get_name(clk_wzrd->clk_in1),
+			0, reg, 1);
+	kfree(clk_name);
+	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul])) {
+		dev_err(&pdev->dev, "unable to register fixed-factor clock\n");
+		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul]);
+		goto err_disable_clk;
+	}
+
+	/* register div */
+	reg = (readl(clk_wzrd->base + WZRD_CLK_CFG_REG(0)) &
+			WZRD_DIVCLK_DIVIDE_MASK) >> WZRD_DIVCLK_DIVIDE_SHIFT;
+	clk_name = kasprintf(GFP_KERNEL, "%s_mul_div", dev_name(&pdev->dev));
+	clk_wzrd->clks_internal[wzrd_clk_mul_div] = clk_register_fixed_factor(
+			&pdev->dev, clk_name,
+			__clk_get_name(clk_wzrd->clks_internal[wzrd_clk_mul]),
+			0, 1, reg);
+	if (IS_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div])) {
+		dev_err(&pdev->dev, "unable to register divider clock\n");
+		ret = PTR_ERR(clk_wzrd->clks_internal[wzrd_clk_mul_div]);
+		goto err_rm_int_clk;
+	}
+
+	/* register div per output */
+	for (i = WZRD_NUM_OUTPUTS - 1; i >= 0 ; i--) {
+		const char *clkout_name;
+		if (of_property_read_string_index(np, "clock-output-names", i,
+						  &clkout_name)) {
+			dev_err(&pdev->dev,
+				"clock output name not specified\n");
+			ret = -EINVAL;
+			goto err_rm_int_clks;
+		}
+		reg = readl(clk_wzrd->base + WZRD_CLK_CFG_REG(2) + i * 12);
+		reg &= WZRD_CLKOUT_DIVIDE_MASK;
+		reg >>= WZRD_CLKOUT_DIVIDE_SHIFT;
+		clk_wzrd->clkout[i] = clk_register_fixed_factor(&pdev->dev,
+				clkout_name, clk_name, 0, 1, reg);
+		if (IS_ERR(clk_wzrd->clkout[i])) {
+			int j;
+			for (j = i + 1; j < WZRD_NUM_OUTPUTS; j++)
+				clk_unregister(clk_wzrd->clkout[j]);
+			dev_err(&pdev->dev,
+				"unable to register divider clock\n");
+			ret = PTR_ERR(clk_wzrd->clkout[i]);
+			goto err_rm_int_clks;
+		}
+	}
+
+	kfree(clk_name);
+
+	clk_wzrd->clk_data.clks = clk_wzrd->clkout;
+	clk_wzrd->clk_data.clk_num = ARRAY_SIZE(clk_wzrd->clkout);
+	of_clk_add_provider(np, of_clk_src_onecell_get, &clk_wzrd->clk_data);
+
+	if (clk_wzrd->speed_grade) {
+		clk_wzrd->nb.notifier_call = clk_wzrd_clk_notifier;
+
+		ret = clk_notifier_register(clk_wzrd->clk_in1,
+					    &clk_wzrd->nb);
+		if (ret)
+			dev_warn(&pdev->dev,
+				 "unable to register clock notifier\n");
+
+		ret = clk_notifier_register(clk_wzrd->axi_clk, &clk_wzrd->nb);
+		if (ret)
+			dev_warn(&pdev->dev,
+				 "unable to register clock notifier\n");
+	}
+
+	return 0;
+
+err_rm_int_clks:
+	clk_unregister(clk_wzrd->clks_internal[1]);
+err_rm_int_clk:
+	kfree(clk_name);
+	clk_unregister(clk_wzrd->clks_internal[0]);
+err_disable_clk:
+	clk_disable_unprepare(clk_wzrd->axi_clk);
+
+	return ret;
+}
+
+static int clk_wzrd_remove(struct platform_device *pdev)
+{
+	int i;
+	struct clk_wzrd *clk_wzrd = platform_get_drvdata(pdev);
+
+	of_clk_del_provider(pdev->dev.of_node);
+
+	for (i = 0; i < WZRD_NUM_OUTPUTS; i++)
+		clk_unregister(clk_wzrd->clkout[i]);
+	for (i = 0; i < wzrd_clk_int_max; i++)
+		clk_unregister(clk_wzrd->clks_internal[i]);
+
+	if (clk_wzrd->speed_grade) {
+		clk_notifier_unregister(clk_wzrd->axi_clk, &clk_wzrd->nb);
+		clk_notifier_unregister(clk_wzrd->clk_in1, &clk_wzrd->nb);
+	}
+
+	clk_disable_unprepare(clk_wzrd->axi_clk);
+
+	return 0;
+}
+
+static const struct of_device_id clk_wzrd_ids[] = {
+	{ .compatible = "xlnx,clocking-wizard" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, clk_wzrd_ids);
+
+static struct platform_driver clk_wzrd_driver = {
+	.driver = {
+		.name = "clk-wizard",
+		.of_match_table = clk_wzrd_ids,
+		.pm = &clk_wzrd_dev_pm_ops,
+	},
+	.probe = clk_wzrd_probe,
+	.remove = clk_wzrd_remove,
+};
+module_platform_driver(clk_wzrd_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Soeren Brinkmann <soren.brinkmann@xilinx.com");
+MODULE_DESCRIPTION("Driver for the Xilinx Clocking Wizard IP core");
diff --git a/drivers/staging/clocking-wizard/dt-binding.txt b/drivers/staging/clocking-wizard/dt-binding.txt
new file mode 100644
index 000000000000..cc3cc5e91440
--- /dev/null
+++ b/drivers/staging/clocking-wizard/dt-binding.txt
@@ -0,0 +1,30 @@
+Binding for Xilinx Clocking Wizard IP Core
+
+This binding uses the common clock binding[1]. Details about the devices can be
+found in the product guide[2].
+
+[1] Documentation/devicetree/bindings/clock/clock-bindings.txt
+[2] Clocking Wizard Product Guide
+http://www.xilinx.com/support/documentation/ip_documentation/clk_wiz/v5_1/pg065-clk-wiz.pdf
+
+Required properties:
+ - compatible: Must be 'xlnx,clocking-wizard'
+ - reg: Base and size of the cores register space
+ - clocks: Handle to input clock
+ - clock-names: Tuple containing 'clk_in1' and 's_axi_aclk'
+ - clock-output-names: Names for the output clocks
+
+Optional properties:
+ - speed-grade: Speed grade of the device
+
+Example:
+	clock-generator@40040000 {
+		reg = <0x40040000 0x1000>;
+		compatible = "xlnx,clocking-wizard";
+		speed-grade = <(-1)>;
+		clock-names = "clk_in1", "s_axi_aclk";
+		clocks = <&clkc 15>, <&clkc 15>;
+		clock-output-names = "clk_out0", "clk_out1", "clk_out2",
+				     "clk_out3", "clk_out4", "clk_out5",
+				     "clk_out6", "clk_out7";
+	};
-- 
2.1.1

Re: [PATCH v2] staging: Add Xilinx Clocking Wizard driver

[Laurent Pinchart]

Hi Sören,

Thank you for the patch.

On Wednesday 01 October 2014 14:02:32 Soren Brinkmann wrote:
> Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
> provides an AXI interface to dynamically reconfigure the clocking
> resources of Xilinx FPGAs.
> 
> Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>
> ---
> Hi Greg, Dan,
> 
> I fixed the things Dan pointed out, please take this v2 instead of the
> original patch.
> 
> Also, don't get caught up too much with style issues around the code that
> registers the clocks. To support set_rate() opeartions this will have to
> change anyway. It doesn't make a lot of sense to spend much time on those
> parts.
> 
> 	Thanks,
> 	Sören
> 
> v2:
>  - implement dev_pm_ops
>  - don't use array for clock inputs
>  - add more information in error output
>  - fix style issues
>  - fix error path
> ---
>  drivers/staging/Kconfig                            |   2 +
>  drivers/staging/Makefile                           |   1 +
>  drivers/staging/clocking-wizard/Kconfig            |   9 +
>  drivers/staging/clocking-wizard/Makefile           |   1 +
>  drivers/staging/clocking-wizard/TODO               |  12 +
>  .../clocking-wizard/clk-xlnx-clock-wizard.c        | 336 ++++++++++++++++++
>  drivers/staging/clocking-wizard/dt-binding.txt     |  30 ++
>  7 files changed, 391 insertions(+)
>  create mode 100644 drivers/staging/clocking-wizard/Kconfig
>  create mode 100644 drivers/staging/clocking-wizard/Makefile
>  create mode 100644 drivers/staging/clocking-wizard/TODO
>  create mode 100644 drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
>  create mode 100644 drivers/staging/clocking-wizard/dt-binding.txt

[snip]

> diff --git a/drivers/staging/clocking-wizard/dt-binding.txt
> b/drivers/staging/clocking-wizard/dt-binding.txt new file mode 100644
> index 000000000000..cc3cc5e91440
> --- /dev/null
> +++ b/drivers/staging/clocking-wizard/dt-binding.txt
> @@ -0,0 +1,30 @@
> +Binding for Xilinx Clocking Wizard IP Core
> +
> +This binding uses the common clock binding[1]. Details about the devices
> can be +found in the product guide[2].
> +
> +[1] Documentation/devicetree/bindings/clock/clock-bindings.txt
> +[2] Clocking Wizard Product Guide
> +http://www.xilinx.com/support/documentation/ip_documentation/clk_wiz/v5_1/p
> g065-clk-wiz.pdf +
> +Required properties:
> + - compatible: Must be 'xlnx,clocking-wizard'
> + - reg: Base and size of the cores register space
> + - clocks: Handle to input clock
> + - clock-names: Tuple containing 'clk_in1' and 's_axi_aclk'
> + - clock-output-names: Names for the output clocks
> +
> +Optional properties:
> + - speed-grade: Speed grade of the device

You should document what values are allowed.

> +
> +Example:
> +	clock-generator@40040000 {
> +		reg = <0x40040000 0x1000>;
> +		compatible = "xlnx,clocking-wizard";
> +		speed-grade = <(-1)>;

Can't we make the speed grade positive ? It looks to me that the - is nowadays 
a dash and not a minus sign, given that higher absolute values of the speed 
grade mean faster devices. The days when the speed grate represented the 
propagation time of signals in ns is long gone I suppose.

> +		clock-names = "clk_in1", "s_axi_aclk";
> +		clocks = <&clkc 15>, <&clkc 15>;
> +		clock-output-names = "clk_out0", "clk_out1", "clk_out2",
> +				     "clk_out3", "clk_out4", "clk_out5",
> +				     "clk_out6", "clk_out7";
> +	};

-- 
Regards,

Laurent Pinchart

Re: [PATCH v2] staging: Add Xilinx Clocking Wizard driver

[Sören Brinkmann]

On Thu, 2014-10-02 at 02:19AM +0300, Laurent Pinchart wrote:
> Hi Sören,
> 
> Thank you for the patch.
> 
> On Wednesday 01 October 2014 14:02:32 Soren Brinkmann wrote:
> > Add a driver for the Xilinx Clocking Wizard soft IP. The clocking wizard
> > provides an AXI interface to dynamically reconfigure the clocking
> > resources of Xilinx FPGAs.
> > 
> > Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>
> > ---
> > Hi Greg, Dan,
> > 
> > I fixed the things Dan pointed out, please take this v2 instead of the
> > original patch.
> > 
> > Also, don't get caught up too much with style issues around the code that
> > registers the clocks. To support set_rate() opeartions this will have to
> > change anyway. It doesn't make a lot of sense to spend much time on those
> > parts.
> > 
> > 	Thanks,
> > 	Sören
> > 
> > v2:
> >  - implement dev_pm_ops
> >  - don't use array for clock inputs
> >  - add more information in error output
> >  - fix style issues
> >  - fix error path
> > ---
> >  drivers/staging/Kconfig                            |   2 +
> >  drivers/staging/Makefile                           |   1 +
> >  drivers/staging/clocking-wizard/Kconfig            |   9 +
> >  drivers/staging/clocking-wizard/Makefile           |   1 +
> >  drivers/staging/clocking-wizard/TODO               |  12 +
> >  .../clocking-wizard/clk-xlnx-clock-wizard.c        | 336 ++++++++++++++++++
> >  drivers/staging/clocking-wizard/dt-binding.txt     |  30 ++
> >  7 files changed, 391 insertions(+)
> >  create mode 100644 drivers/staging/clocking-wizard/Kconfig
> >  create mode 100644 drivers/staging/clocking-wizard/Makefile
> >  create mode 100644 drivers/staging/clocking-wizard/TODO
> >  create mode 100644 drivers/staging/clocking-wizard/clk-xlnx-clock-wizard.c
> >  create mode 100644 drivers/staging/clocking-wizard/dt-binding.txt
> 
> [snip]
> 
> > diff --git a/drivers/staging/clocking-wizard/dt-binding.txt
> > b/drivers/staging/clocking-wizard/dt-binding.txt new file mode 100644
> > index 000000000000..cc3cc5e91440
> > --- /dev/null
> > +++ b/drivers/staging/clocking-wizard/dt-binding.txt
> > @@ -0,0 +1,30 @@
> > +Binding for Xilinx Clocking Wizard IP Core
> > +
> > +This binding uses the common clock binding[1]. Details about the devices
> > can be +found in the product guide[2].
> > +
> > +[1] Documentation/devicetree/bindings/clock/clock-bindings.txt
> > +[2] Clocking Wizard Product Guide
> > +http://www.xilinx.com/support/documentation/ip_documentation/clk_wiz/v5_1/p
> > g065-clk-wiz.pdf +
> > +Required properties:
> > + - compatible: Must be 'xlnx,clocking-wizard'
> > + - reg: Base and size of the cores register space
> > + - clocks: Handle to input clock
> > + - clock-names: Tuple containing 'clk_in1' and 's_axi_aclk'
> > + - clock-output-names: Names for the output clocks
> > +
> > +Optional properties:
> > + - speed-grade: Speed grade of the device
> 
> You should document what values are allowed.

Will do.

> > +
> > +Example:
> > +	clock-generator@40040000 {
> > +		reg = <0x40040000 0x1000>;
> > +		compatible = "xlnx,clocking-wizard";
> > +		speed-grade = <(-1)>;
> 
> Can't we make the speed grade positive ? It looks to me that the - is nowadays 
> a dash and not a minus sign, given that higher absolute values of the speed 
> grade mean faster devices. The days when the speed grate represented the 
> propagation time of signals in ns is long gone I suppose.

Fine with me.

	Thanks,
	Sören