* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-06-29 18:32 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Haojian Zhuang, Eric Miao
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland,
Robert Jarzmik
As the RFC posted in [1] didn't meet an unrivaled success for
review, I'm posting this serie for PXA27x transition to clock
framework.
This transition is needed :
- to enable device-tree drivers port, as clocks are needed almost
everywhere
- to enable the long term multi-platform kernel to support PXA
As I had said before, this serie aims at :
- keeping legacy platforms working (ie. without device-tree)
- enable PXA27x to work with a device-tree kernel, and hence
open the way to drivers conversion
- be robust enough to support pxa25x and pxa3xx later inclusion
with almost no change to clk-pxa-dt.c.
As this serie is holding the rest of the device-tree drivers
port, I'd like it to be reviewed, even it's an old unsexy
platform.
Cheers.
--
Robert
[1]: http://www.spinics.net/lists/arm-kernel/msg337521.html
Robert Jarzmik (4):
clk: add pxa27x clock drivers
dts: add devicetree bindings for pxa27x clocks
arm: pxa: Transition pxa27x to clk framework
clk: dts: document pxa27x clock binding
.../devicetree/bindings/clock/pxa-clock.txt | 32 ++
arch/arm/Kconfig | 1 +
arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++-
arch/arm/mach-pxa/Makefile | 8 +-
arch/arm/mach-pxa/pxa27x.c | 190 +-----------
drivers/clk/Makefile | 1 +
drivers/clk/pxa/Makefile | 4 +
drivers/clk/pxa/clk-pxa-dt.c | 76 +++++
drivers/clk/pxa/clk-pxa27x.c | 324 +++++++++++++++++++++
drivers/clk/pxa/clk-pxa2xx.c | 74 +++++
drivers/clk/pxa/clk-pxa2xx.h | 47 +++
include/dt-bindings/clock/pxa2xx-clock.h | 45 +++
12 files changed, 757 insertions(+), 179 deletions(-)
create mode 100644 Documentation/devicetree/bindings/clock/pxa-clock.txt
create mode 100644 drivers/clk/pxa/Makefile
create mode 100644 drivers/clk/pxa/clk-pxa-dt.c
create mode 100644 drivers/clk/pxa/clk-pxa27x.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.h
create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
--
2.0.0.rc2
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-06-29 18:32 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: linux-arm-kernel
As the RFC posted in [1] didn't meet an unrivaled success for
review, I'm posting this serie for PXA27x transition to clock
framework.
This transition is needed :
- to enable device-tree drivers port, as clocks are needed almost
everywhere
- to enable the long term multi-platform kernel to support PXA
As I had said before, this serie aims at :
- keeping legacy platforms working (ie. without device-tree)
- enable PXA27x to work with a device-tree kernel, and hence
open the way to drivers conversion
- be robust enough to support pxa25x and pxa3xx later inclusion
with almost no change to clk-pxa-dt.c.
As this serie is holding the rest of the device-tree drivers
port, I'd like it to be reviewed, even it's an old unsexy
platform.
Cheers.
--
Robert
[1]: http://www.spinics.net/lists/arm-kernel/msg337521.html
Robert Jarzmik (4):
clk: add pxa27x clock drivers
dts: add devicetree bindings for pxa27x clocks
arm: pxa: Transition pxa27x to clk framework
clk: dts: document pxa27x clock binding
.../devicetree/bindings/clock/pxa-clock.txt | 32 ++
arch/arm/Kconfig | 1 +
arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++-
arch/arm/mach-pxa/Makefile | 8 +-
arch/arm/mach-pxa/pxa27x.c | 190 +-----------
drivers/clk/Makefile | 1 +
drivers/clk/pxa/Makefile | 4 +
drivers/clk/pxa/clk-pxa-dt.c | 76 +++++
drivers/clk/pxa/clk-pxa27x.c | 324 +++++++++++++++++++++
drivers/clk/pxa/clk-pxa2xx.c | 74 +++++
drivers/clk/pxa/clk-pxa2xx.h | 47 +++
include/dt-bindings/clock/pxa2xx-clock.h | 45 +++
12 files changed, 757 insertions(+), 179 deletions(-)
create mode 100644 Documentation/devicetree/bindings/clock/pxa-clock.txt
create mode 100644 drivers/clk/pxa/Makefile
create mode 100644 drivers/clk/pxa/clk-pxa-dt.c
create mode 100644 drivers/clk/pxa/clk-pxa27x.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.h
create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
--
2.0.0.rc2
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 1/4] clk: add pxa27x clock drivers
2014-06-29 18:32 ` Robert Jarzmik
@ 2014-06-29 18:32 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Haojian Zhuang, Eric Miao
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland,
Robert Jarzmik
Move pxa27x clock drivers from arch/arm/mach-pxa to driver/clk.
In the move :
- convert to new clock framework legacy clocks
- provide clocks as before for platform data based boards
- provide clocks through devicetree with clk-pxa-dt
This is the preliminary step in the conversion. The remaining steps are
:
- migrate pxa25x and pxa3xx
- once PXA is fully converted to device tree, if that happens,
clk-pxa2* and clk-pxa3* should only hold the core clocks which cannot
be described in devicetree.
Signed-off-by: Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>
---
drivers/clk/Makefile | 1 +
drivers/clk/pxa/Makefile | 4 +
drivers/clk/pxa/clk-pxa-dt.c | 76 ++++++++++
drivers/clk/pxa/clk-pxa27x.c | 324 +++++++++++++++++++++++++++++++++++++++++++
drivers/clk/pxa/clk-pxa2xx.c | 74 ++++++++++
drivers/clk/pxa/clk-pxa2xx.h | 47 +++++++
6 files changed, 526 insertions(+)
create mode 100644 drivers/clk/pxa/Makefile
create mode 100644 drivers/clk/pxa/clk-pxa-dt.c
create mode 100644 drivers/clk/pxa/clk-pxa27x.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.h
diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
index 567f102..40390ea 100644
--- a/drivers/clk/Makefile
+++ b/drivers/clk/Makefile
@@ -43,6 +43,7 @@ obj-$(CONFIG_ARCH_MMP) += mmp/
endif
obj-$(CONFIG_PLAT_ORION) += mvebu/
obj-$(CONFIG_ARCH_MXS) += mxs/
+obj-$(CONFIG_ARCH_PXA) += pxa/
obj-$(CONFIG_COMMON_CLK_QCOM) += qcom/
obj-$(CONFIG_ARCH_ROCKCHIP) += rockchip/
obj-$(CONFIG_COMMON_CLK_SAMSUNG) += samsung/
diff --git a/drivers/clk/pxa/Makefile b/drivers/clk/pxa/Makefile
new file mode 100644
index 0000000..65ff461
--- /dev/null
+++ b/drivers/clk/pxa/Makefile
@@ -0,0 +1,4 @@
+obj-y += clk-pxa-dt.o
+obj-$(CONFIG_PXA25x) += clk-pxa2xx.o clk-pxa25x.o
+obj-$(CONFIG_PXA27x) += clk-pxa2xx.o clk-pxa27x.o
+obj-$(CONFIG_PXA3xx) += clk-pxa3xx.o
diff --git a/drivers/clk/pxa/clk-pxa-dt.c b/drivers/clk/pxa/clk-pxa-dt.c
new file mode 100644
index 0000000..530e75b
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa-dt.c
@@ -0,0 +1,76 @@
+/*
+ * Marvell PXA2xx family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <mach/pxa2xx-regs.h>
+
+#define PXA2XX_MAX_CLOCKS 32
+
+static DEFINE_SPINLOCK(cken_lock);
+
+static void __init pxa2xx_clocks_init(struct device_node *np)
+{
+ struct clk *clk;
+ struct clk_onecell_data *onecell_data = NULL;
+ const char *name;
+ const char *parent_name;
+ int i, ret;
+ u32 clkidx;
+ void __iomem *reg;
+
+ ret = -ENOMEM;
+ onecell_data = kmalloc(sizeof(*onecell_data), GFP_KERNEL);
+ if (!onecell_data)
+ goto err_mem;
+ onecell_data->clks = kcalloc(PXA2XX_MAX_CLOCKS, sizeof(struct clk *),
+ GFP_KERNEL);
+ if (!onecell_data->clks)
+ goto err_mem;
+ onecell_data->clk_num = PXA2XX_MAX_CLOCKS;
+
+ reg = of_iomap(np, 0);
+ if (!reg)
+ return;
+
+ for (i = 0; i < PXA2XX_MAX_CLOCKS; i++) {
+ ret = of_property_read_string_index(np, "clock-output-names",
+ i, &name);
+ if (ret < 0 || strlen(name) == 0)
+ continue;
+
+ parent_name = of_clk_get_parent_name(np, i);
+ ret = of_property_read_u32_index(np, "clock-indices", i,
+ &clkidx);
+ if (parent_name == NULL || ret < 0)
+ break;
+ if (clkidx >= PXA2XX_MAX_CLOCKS) {
+ pr_err("%s: invalid clock %s %s index %u)\n",
+ __func__, np->name, name, clkidx);
+ continue;
+ }
+
+ clk = clk_register_gate(NULL, name, parent_name, 0,
+ reg, clkidx, 0, &cken_lock);
+ if (!IS_ERR(clk))
+ onecell_data->clks[i] = clk;
+ }
+
+ of_clk_add_provider(np, of_clk_src_onecell_get, onecell_data);
+ return;
+err_mem:
+ if (onecell_data)
+ kfree(onecell_data->clks);
+ kfree(onecell_data);
+}
+CLK_OF_DECLARE(pxa2xx_clks, "marvell,pxa-clocks", pxa2xx_clocks_init);
diff --git a/drivers/clk/pxa/clk-pxa27x.c b/drivers/clk/pxa/clk-pxa27x.c
new file mode 100644
index 0000000..5e1e172
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa27x.c
@@ -0,0 +1,324 @@
+/*
+ * Marvell PXA27x family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * Heavily inspired from former arch/arm/mach-pxa/clock.c.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * For non-devicetree platforms. Once pxa is fully converted to devicetree, this
+ * should go away.
+ */
+#include <linux/clk-provider.h>
+#include <mach/pxa2xx-regs.h>
+#include <linux/io.h>
+#include <linux/clk.h>
+#include <linux/clkdev.h>
+#include <linux/of.h>
+
+#include "clk-pxa2xx.h"
+
+static DEFINE_SPINLOCK(cken_lock);
+
+/* Crystal clock: 13MHz */
+#define BASE_CLK 13000000
+
+static unsigned long get_run_clock(unsigned long parent_rate)
+{
+ unsigned long ccsr = CCSR;
+ unsigned int l = ccsr & 0x1f;
+
+ return parent_rate * l;
+}
+
+static unsigned long get_turbo_clock(unsigned long parent_rate)
+{
+ unsigned long ccsr = CCSR;
+ unsigned int n2 = (ccsr >> 7) & 0xf;
+
+ return (get_run_clock(parent_rate) * n2) / 2;
+}
+
+static unsigned long get_halfturbo_clock(unsigned long parent_rate)
+{
+ return get_turbo_clock(parent_rate) / 2;
+}
+
+static unsigned long get_cpu_core_clock(unsigned long parent_rate)
+{
+ unsigned long clkcfg;
+ unsigned int t, ht, b;
+
+ asm("mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg));
+ t = clkcfg & (1 << 0);
+ ht = clkcfg & (1 << 2);
+ b = clkcfg & (1 << 3);
+
+ if (ht)
+ return get_halfturbo_clock(parent_rate);
+ if (t)
+ return get_turbo_clock(parent_rate);
+ return get_run_clock(parent_rate);
+}
+
+static unsigned long get_sysbus_clock(unsigned long parent_rate)
+{
+ unsigned long clkcfg;
+ unsigned int t, ht, b;
+
+ asm("mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg));
+ t = clkcfg & (1 << 0);
+ ht = clkcfg & (1 << 2);
+ b = clkcfg & (1 << 3);
+
+ return b ? get_run_clock(parent_rate) : get_run_clock(parent_rate) / 2;
+}
+
+static unsigned long get_memory_clock(unsigned long parent_rate)
+{
+ int cccr_a = CCCR & (1 << 25);
+ unsigned long ccsr = CCSR;
+ unsigned int l = ccsr & 0x1f;
+
+ cccr_a = CCCR & (1 << 25);
+ if (cccr_a)
+ return get_sysbus_clock(parent_rate);
+ if (l <= 10)
+ return get_run_clock(parent_rate);
+ if (l <= 20)
+ return get_run_clock(parent_rate) / 2;
+ return get_run_clock(parent_rate) / 4;
+};
+
+static unsigned long get_lcd_clock(unsigned long parent_rate)
+{
+ unsigned long ccsr = CCSR;
+ unsigned int l = ccsr & 0x1f;
+ int k = (l <= 7) ? 1 : (l <= 16) ? 2 : 4;
+
+ return get_run_clock(parent_rate) / k;
+}
+
+
+/*
+ * Get the clock frequency as reflected by CCSR and the turbo flag.
+ * We assume these values have been applied via a fcs.
+ * If info is not 0 we also display the current settings.
+ */
+unsigned int pxa27x_get_clk_frequency_khz(int info)
+{
+ unsigned long core_clk, run_mode, turbo_mode, mem_clk, sys_clk;
+
+ core_clk = get_cpu_core_clock(BASE_CLK);
+ run_mode = get_run_clock(BASE_CLK);
+ turbo_mode = get_turbo_clock(BASE_CLK);
+ mem_clk = get_memory_clock(BASE_CLK);
+ sys_clk = get_sysbus_clock(BASE_CLK);
+
+ if (info) {
+ pr_info("Run Mode clock: %ld.%02ldMHz\n",
+ run_mode / 1000000, (run_mode % 1000000) / 10000);
+ pr_info("Turbo Mode clock: %ld.%02ldMHz\n",
+ turbo_mode / 1000000, (turbo_mode % 1000000) / 10000);
+ pr_info("Memory clock: %ld.%02ldMHz\n",
+ mem_clk / 1000000, (mem_clk % 1000000) / 10000);
+ pr_info("System bus clock: %ld.%02ldMHz\n",
+ sys_clk / 1000000, (sys_clk % 1000000) / 10000);
+ }
+
+ return core_clk;
+}
+
+/*
+ * Return the current mem clock frequency as reflected by CCCR[A], B, and L
+ */
+static unsigned long clk_pxa27x_mem_getrate(struct clk_hw *hw,
+ unsigned long parent_rate)
+{
+ return get_memory_clock(BASE_CLK);
+}
+
+static const struct clk_ops clk_pxa27x_mem_ops = {
+ .recalc_rate = clk_pxa27x_mem_getrate,
+};
+
+static unsigned long clk_pxa27x_lcd_getrate(struct clk_hw *hw,
+ unsigned long parent_rate)
+{
+ return get_lcd_clock(BASE_CLK);
+}
+
+static const struct clk_ops clk_pxa27x_lcd_ops = {
+ .recalc_rate = clk_pxa27x_lcd_getrate,
+};
+
+static struct pxa27x_clocks_fixed_cken pxa27x_without_dt_clocks[] __initdata = {
+ PXA2_FIXED_RATE("pxa2xx-uart.0", NULL, FFUART, 14857000, 1),
+ PXA2_FIXED_RATE("pxa2xx-uart.1", NULL, BTUART, 14857000, 1),
+ PXA2_FIXED_RATE("pxa2xx-uart.2", NULL, STUART, 14857000, 1),
+ PXA2_FIXED_RATE(NULL, "UARTCLK", STUART, 14857000, 1),
+ PXA2_FIXED_RATE("pxa2xx-i2s", NULL, I2S, 14682000, 0),
+ PXA2_FIXED_RATE("pxa2xx-i2c.0", NULL, I2C, 32842000, 0),
+ PXA2_FIXED_RATE("pxa27x-udc", NULL, USB, 48000000, 5),
+ PXA2_FIXED_RATE("pxa2xx-mci.0", NULL, MMC, 19500000, 0),
+ PXA2_FIXED_RATE("pxa2xx-ir", "FICPCLK", FICP, 48000000, 0),
+ PXA2_FIXED_RATE("pxa27x-ohci", NULL, USBHOST, 48000000, 0),
+ PXA2_FIXED_RATE("pxa2xx-i2c.1", NULL, PWRI2C, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-keypad", NULL, KEYPAD, 32768, 0),
+ PXA2_FIXED_RATE("pxa27x-ssp.0", NULL, SSP1, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-ssp.1", NULL, SSP2, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-ssp.2", NULL, SSP3, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-pwm.0", NULL, PWM0, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-pwm.1", NULL, PWM1, 13000000, 0),
+ PXA2_FIXED_RATE(NULL, "AC97CLK", AC97, 24576000, 0),
+ PXA2_FIXED_RATE(NULL, "AC97CONFCLK", AC97CONF, 24576000, 0),
+ PXA2_FIXED_RATE(NULL, "MSLCLK", MSL, 48000000, 0),
+ PXA2_FIXED_RATE(NULL, "USIMCLK", USIM, 48000000, 0),
+ PXA2_FIXED_RATE(NULL, "MSTKCLK", MEMSTK, 19500000, 0),
+ PXA2_FIXED_RATE(NULL, "IMCLK", IM, 0, 0),
+ PXA2_FIXED_RATE_AO("pxa27x-memc", "MEMCLK", MEMC, 0, 0),
+};
+
+static struct pxa27x_clocks_var_rate pxa27x_var_rate_clocks[] __initdata = {
+ PXA2_VAR_RATE("pxa2xx-fb", true, LCD, &clk_pxa27x_lcd_ops),
+ PXA2_VAR_RATE("pxa27x-camera.0", true, CAMERA, &clk_pxa27x_lcd_ops),
+ PXA2_VAR_RATE("pxa2xx-pcmcia", false, MEMC, &clk_pxa27x_mem_ops),
+};
+
+static int __init pxa27x_clocks_init(void)
+{
+ int i;
+ unsigned long rate, flags;
+ struct clk *clk_parent, *clk;
+ struct pxa27x_clocks_fixed_cken *fclock;
+ struct pxa27x_clocks_var_rate *vclock;
+ char parent_name[80], *name;
+
+ /*
+ * Once device-tree conversion is complete :
+ * if (of_have_populated_dt())
+ * return 0;
+ */
+ for (i = 0; i < ARRAY_SIZE(pxa27x_without_dt_clocks); i++) {
+ fclock = &pxa27x_without_dt_clocks[i];
+ clk_parent = NULL;
+ rate = fclock->rate;
+ snprintf(parent_name, sizeof(parent_name), "clk-%lu", rate);
+ flags = CLK_GET_RATE_NOCACHE | CLK_IS_ROOT;
+ clk_parent = clk_register_fixed_rate(NULL, parent_name, NULL,
+ flags, rate);
+ name = fclock->name ? fclock->name : fclock->con_id;
+ flags = fclock->always_on ? CLK_IGNORE_UNUSED : 0;
+ clk = clk_register_gate(NULL, name, parent_name, flags,
+ (void *)&CKEN, fclock->cken, 0,
+ &cken_lock);
+ if (!IS_ERR(clk))
+ clk_register_clkdev(clk, fclock->con_id, fclock->name);
+ }
+
+ for (i = 0; i < ARRAY_SIZE(pxa27x_var_rate_clocks); i++) {
+ vclock = &pxa27x_var_rate_clocks[i];
+ flags = CLK_GET_RATE_NOCACHE;
+ if (vclock->has_cken) {
+ snprintf(parent_name, sizeof(parent_name), "clk-%s",
+ vclock->name);
+ clk_parent = clk_pxa2xx_register_dummy(parent_name,
+ vclock->clk_ops);
+ clk = clk_register_gate(NULL, vclock->name, parent_name,
+ flags, (void *)&CKEN,
+ vclock->cken, 0, &cken_lock);
+ } else {
+ clk = clk_pxa2xx_register_dummy(vclock->name,
+ vclock->clk_ops);
+ }
+ if (!IS_ERR(clk))
+ clk_register_clkdev(clk, NULL, vclock->name);
+ }
+ return 0;
+}
+
+postcore_initcall(pxa27x_clocks_init);
+
+struct clk_pxa27x_core {
+ struct clk_hw hw;
+ unsigned long (*get_rate)(unsigned long parent_rate);
+};
+
+static unsigned long clk_pxa27x_ops_get_rate(struct clk_hw *hw,
+ unsigned long parent_rate)
+{
+ struct clk_pxa27x_core *clk_core =
+ container_of(hw, struct clk_pxa27x_core, hw);
+
+ return clk_core->get_rate(parent_rate);
+}
+
+const struct clk_ops clk_pxa27x_ops = {
+ .recalc_rate = clk_pxa27x_ops_get_rate,
+};
+
+static struct clk __init *clk_register_pxa27x_core(const char *name,
+ const char *parent_name,
+ unsigned long (*get_rate)(unsigned long))
+{
+ struct clk_pxa27x_core *clk_core;
+ struct clk_init_data init;
+ struct clk *clk;
+
+ clk_core = kzalloc(sizeof(*clk_core), GFP_KERNEL);
+ if (!clk_core) {
+ pr_err("%s: could not allocate pxa27x_core clock\n", __func__);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ init.name = name;
+ init.ops = &clk_pxa27x_ops;
+ init.flags = CLK_IS_BASIC | CLK_GET_RATE_NOCACHE;
+ init.parent_names = (parent_name ? &parent_name : NULL);
+ init.num_parents = (parent_name ? 1 : 0);
+
+ clk_core->get_rate = get_rate;
+ clk_core->hw.init = &init;
+ clk = clk_register(NULL, &clk_core->hw);
+ return clk;
+}
+
+#define NB_CORE_CLOCKS 7
+static struct clk *core_clks[NB_CORE_CLOCKS];
+static struct clk_onecell_data onecell_data = {
+ .clks = core_clks,
+ .clk_num = NB_CORE_CLOCKS,
+};
+
+static void __init pxa27x_core_clocks_init(struct device_node *np)
+{
+ const char *pname;
+
+ pname = of_clk_get_parent_name(np, 0);
+ onecell_data.clks[0] =
+ clk_register_pxa27x_core("run mode", pname, get_run_clock);
+ onecell_data.clks[1] =
+ clk_register_pxa27x_core("turbo mode", pname, get_turbo_clock);
+ onecell_data.clks[2] =
+ clk_register_pxa27x_core("halt-turbo mode", pname,
+ get_halfturbo_clock);
+ onecell_data.clks[3] =
+ clk_register_pxa27x_core("cpu core", pname,
+ get_cpu_core_clock);
+ onecell_data.clks[4] =
+ clk_register_pxa27x_core("system bus", pname,
+ get_sysbus_clock);
+ onecell_data.clks[5] =
+ clk_register_pxa27x_core("memory", pname,
+ get_memory_clock);
+ onecell_data.clks[5] =
+ clk_register_pxa27x_core("lcd", pname,
+ get_lcd_clock);
+
+ of_clk_add_provider(np, of_clk_src_onecell_get, &onecell_data);
+}
+CLK_OF_DECLARE(pxa2xx_clks, "marvell,pxa270-core-clocks",
+ pxa27x_core_clocks_init);
diff --git a/drivers/clk/pxa/clk-pxa2xx.c b/drivers/clk/pxa/clk-pxa2xx.c
new file mode 100644
index 0000000..6e6d657
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa2xx.c
@@ -0,0 +1,74 @@
+/*
+ * Marvell PXA2xx family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/syscore_ops.h>
+#include <mach/pxa2xx-regs.h>
+
+struct dummy_clk {
+ struct clk_hw hw;
+};
+
+struct clk * __init clk_pxa2xx_register_dummy(const char *name,
+ const struct clk_ops *ops)
+{
+ struct clk_init_data init;
+ struct dummy_clk *clock;
+ struct clk *clk;
+
+ clock = kzalloc(sizeof(*clock), GFP_KERNEL);
+ if (!clock) {
+ pr_err("%s: failed to allocate PXA2xx clock.\n", __func__);
+ return ERR_PTR(-ENOMEM);
+ }
+ init.name = name;
+ init.ops = ops;
+ init.parent_names = NULL;
+ init.num_parents = 0;
+ init.flags = CLK_IS_BASIC | CLK_GET_RATE_NOCACHE;
+ clock->hw.init = &init;
+ clk = clk_register(NULL, &clock->hw);
+ if (IS_ERR(clk))
+ kfree(clock);
+ return clk;
+}
+
+
+#ifdef CONFIG_PM
+static uint32_t saved_cken;
+
+static int pxa2xx_clock_suspend(void)
+{
+ saved_cken = CKEN;
+ return 0;
+}
+
+static void pxa2xx_clock_resume(void)
+{
+ CKEN = saved_cken;
+}
+#else
+#define pxa2xx_clock_suspend NULL
+#define pxa2xx_clock_resume NULL
+#endif
+
+static struct syscore_ops pxa2xx_clock_syscore_ops = {
+ .suspend = pxa2xx_clock_suspend,
+ .resume = pxa2xx_clock_resume,
+};
+
+static int __init pxa2xx_clocks_init(void)
+{
+ register_syscore_ops(&pxa2xx_clock_syscore_ops);
+ return 0;
+}
+
+postcore_initcall(pxa2xx_clocks_init);
diff --git a/drivers/clk/pxa/clk-pxa2xx.h b/drivers/clk/pxa/clk-pxa2xx.h
new file mode 100644
index 0000000..79f62af
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa2xx.h
@@ -0,0 +1,47 @@
+/*
+ * Marvell PXA27x family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * For non-devicetree platforms. Once pxa is fully converted to devicetree, this
+ * should go away.
+ */
+
+#ifndef _CLK_PXA2XX_H_
+#define _CLK_PXA2XX_H_
+
+#define PXA2_VAR_RATE(_name, _has_cken, _cken, _clk_ops) \
+ { .name = _name, .has_cken = _has_cken, .cken = CKEN_ ## _cken, \
+ .clk_ops = _clk_ops }
+
+struct pxa27x_clocks_var_rate {
+ char *name;
+ bool has_cken;
+ unsigned int cken;
+ const struct clk_ops *clk_ops;
+};
+
+#define PXA2_FIXED_RATE(_name, _con_id, _cken, _rate, _delay) \
+ { .name = _name, .con_id = _con_id, .cken = CKEN_ ## _cken, \
+ .rate = _rate, .delay = _delay, .always_on = false, }
+#define PXA2_FIXED_RATE_AO(_name, _con_id, _cken, _rate, _delay) \
+ { .name = _name, .con_id = _con_id, .cken = CKEN_ ## _cken, \
+ .rate = _rate, .delay = _delay, .always_on = true, }
+
+struct pxa27x_clocks_fixed_cken {
+ char *name;
+ char *con_id;
+ unsigned int cken;
+ unsigned long rate;
+ int delay;
+ bool always_on;
+};
+
+extern struct clk * __init clk_pxa2xx_register_dummy(const char *name,
+ const struct clk_ops *ops);
+
+#endif
--
2.0.0.rc2
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [PATCH 1/4] clk: add pxa27x clock drivers
@ 2014-06-29 18:32 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: linux-arm-kernel
Move pxa27x clock drivers from arch/arm/mach-pxa to driver/clk.
In the move :
- convert to new clock framework legacy clocks
- provide clocks as before for platform data based boards
- provide clocks through devicetree with clk-pxa-dt
This is the preliminary step in the conversion. The remaining steps are
:
- migrate pxa25x and pxa3xx
- once PXA is fully converted to device tree, if that happens,
clk-pxa2* and clk-pxa3* should only hold the core clocks which cannot
be described in devicetree.
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
---
drivers/clk/Makefile | 1 +
drivers/clk/pxa/Makefile | 4 +
drivers/clk/pxa/clk-pxa-dt.c | 76 ++++++++++
drivers/clk/pxa/clk-pxa27x.c | 324 +++++++++++++++++++++++++++++++++++++++++++
drivers/clk/pxa/clk-pxa2xx.c | 74 ++++++++++
drivers/clk/pxa/clk-pxa2xx.h | 47 +++++++
6 files changed, 526 insertions(+)
create mode 100644 drivers/clk/pxa/Makefile
create mode 100644 drivers/clk/pxa/clk-pxa-dt.c
create mode 100644 drivers/clk/pxa/clk-pxa27x.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.c
create mode 100644 drivers/clk/pxa/clk-pxa2xx.h
diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
index 567f102..40390ea 100644
--- a/drivers/clk/Makefile
+++ b/drivers/clk/Makefile
@@ -43,6 +43,7 @@ obj-$(CONFIG_ARCH_MMP) += mmp/
endif
obj-$(CONFIG_PLAT_ORION) += mvebu/
obj-$(CONFIG_ARCH_MXS) += mxs/
+obj-$(CONFIG_ARCH_PXA) += pxa/
obj-$(CONFIG_COMMON_CLK_QCOM) += qcom/
obj-$(CONFIG_ARCH_ROCKCHIP) += rockchip/
obj-$(CONFIG_COMMON_CLK_SAMSUNG) += samsung/
diff --git a/drivers/clk/pxa/Makefile b/drivers/clk/pxa/Makefile
new file mode 100644
index 0000000..65ff461
--- /dev/null
+++ b/drivers/clk/pxa/Makefile
@@ -0,0 +1,4 @@
+obj-y += clk-pxa-dt.o
+obj-$(CONFIG_PXA25x) += clk-pxa2xx.o clk-pxa25x.o
+obj-$(CONFIG_PXA27x) += clk-pxa2xx.o clk-pxa27x.o
+obj-$(CONFIG_PXA3xx) += clk-pxa3xx.o
diff --git a/drivers/clk/pxa/clk-pxa-dt.c b/drivers/clk/pxa/clk-pxa-dt.c
new file mode 100644
index 0000000..530e75b
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa-dt.c
@@ -0,0 +1,76 @@
+/*
+ * Marvell PXA2xx family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <mach/pxa2xx-regs.h>
+
+#define PXA2XX_MAX_CLOCKS 32
+
+static DEFINE_SPINLOCK(cken_lock);
+
+static void __init pxa2xx_clocks_init(struct device_node *np)
+{
+ struct clk *clk;
+ struct clk_onecell_data *onecell_data = NULL;
+ const char *name;
+ const char *parent_name;
+ int i, ret;
+ u32 clkidx;
+ void __iomem *reg;
+
+ ret = -ENOMEM;
+ onecell_data = kmalloc(sizeof(*onecell_data), GFP_KERNEL);
+ if (!onecell_data)
+ goto err_mem;
+ onecell_data->clks = kcalloc(PXA2XX_MAX_CLOCKS, sizeof(struct clk *),
+ GFP_KERNEL);
+ if (!onecell_data->clks)
+ goto err_mem;
+ onecell_data->clk_num = PXA2XX_MAX_CLOCKS;
+
+ reg = of_iomap(np, 0);
+ if (!reg)
+ return;
+
+ for (i = 0; i < PXA2XX_MAX_CLOCKS; i++) {
+ ret = of_property_read_string_index(np, "clock-output-names",
+ i, &name);
+ if (ret < 0 || strlen(name) == 0)
+ continue;
+
+ parent_name = of_clk_get_parent_name(np, i);
+ ret = of_property_read_u32_index(np, "clock-indices", i,
+ &clkidx);
+ if (parent_name == NULL || ret < 0)
+ break;
+ if (clkidx >= PXA2XX_MAX_CLOCKS) {
+ pr_err("%s: invalid clock %s %s index %u)\n",
+ __func__, np->name, name, clkidx);
+ continue;
+ }
+
+ clk = clk_register_gate(NULL, name, parent_name, 0,
+ reg, clkidx, 0, &cken_lock);
+ if (!IS_ERR(clk))
+ onecell_data->clks[i] = clk;
+ }
+
+ of_clk_add_provider(np, of_clk_src_onecell_get, onecell_data);
+ return;
+err_mem:
+ if (onecell_data)
+ kfree(onecell_data->clks);
+ kfree(onecell_data);
+}
+CLK_OF_DECLARE(pxa2xx_clks, "marvell,pxa-clocks", pxa2xx_clocks_init);
diff --git a/drivers/clk/pxa/clk-pxa27x.c b/drivers/clk/pxa/clk-pxa27x.c
new file mode 100644
index 0000000..5e1e172
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa27x.c
@@ -0,0 +1,324 @@
+/*
+ * Marvell PXA27x family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * Heavily inspired from former arch/arm/mach-pxa/clock.c.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * For non-devicetree platforms. Once pxa is fully converted to devicetree, this
+ * should go away.
+ */
+#include <linux/clk-provider.h>
+#include <mach/pxa2xx-regs.h>
+#include <linux/io.h>
+#include <linux/clk.h>
+#include <linux/clkdev.h>
+#include <linux/of.h>
+
+#include "clk-pxa2xx.h"
+
+static DEFINE_SPINLOCK(cken_lock);
+
+/* Crystal clock: 13MHz */
+#define BASE_CLK 13000000
+
+static unsigned long get_run_clock(unsigned long parent_rate)
+{
+ unsigned long ccsr = CCSR;
+ unsigned int l = ccsr & 0x1f;
+
+ return parent_rate * l;
+}
+
+static unsigned long get_turbo_clock(unsigned long parent_rate)
+{
+ unsigned long ccsr = CCSR;
+ unsigned int n2 = (ccsr >> 7) & 0xf;
+
+ return (get_run_clock(parent_rate) * n2) / 2;
+}
+
+static unsigned long get_halfturbo_clock(unsigned long parent_rate)
+{
+ return get_turbo_clock(parent_rate) / 2;
+}
+
+static unsigned long get_cpu_core_clock(unsigned long parent_rate)
+{
+ unsigned long clkcfg;
+ unsigned int t, ht, b;
+
+ asm("mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg));
+ t = clkcfg & (1 << 0);
+ ht = clkcfg & (1 << 2);
+ b = clkcfg & (1 << 3);
+
+ if (ht)
+ return get_halfturbo_clock(parent_rate);
+ if (t)
+ return get_turbo_clock(parent_rate);
+ return get_run_clock(parent_rate);
+}
+
+static unsigned long get_sysbus_clock(unsigned long parent_rate)
+{
+ unsigned long clkcfg;
+ unsigned int t, ht, b;
+
+ asm("mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg));
+ t = clkcfg & (1 << 0);
+ ht = clkcfg & (1 << 2);
+ b = clkcfg & (1 << 3);
+
+ return b ? get_run_clock(parent_rate) : get_run_clock(parent_rate) / 2;
+}
+
+static unsigned long get_memory_clock(unsigned long parent_rate)
+{
+ int cccr_a = CCCR & (1 << 25);
+ unsigned long ccsr = CCSR;
+ unsigned int l = ccsr & 0x1f;
+
+ cccr_a = CCCR & (1 << 25);
+ if (cccr_a)
+ return get_sysbus_clock(parent_rate);
+ if (l <= 10)
+ return get_run_clock(parent_rate);
+ if (l <= 20)
+ return get_run_clock(parent_rate) / 2;
+ return get_run_clock(parent_rate) / 4;
+};
+
+static unsigned long get_lcd_clock(unsigned long parent_rate)
+{
+ unsigned long ccsr = CCSR;
+ unsigned int l = ccsr & 0x1f;
+ int k = (l <= 7) ? 1 : (l <= 16) ? 2 : 4;
+
+ return get_run_clock(parent_rate) / k;
+}
+
+
+/*
+ * Get the clock frequency as reflected by CCSR and the turbo flag.
+ * We assume these values have been applied via a fcs.
+ * If info is not 0 we also display the current settings.
+ */
+unsigned int pxa27x_get_clk_frequency_khz(int info)
+{
+ unsigned long core_clk, run_mode, turbo_mode, mem_clk, sys_clk;
+
+ core_clk = get_cpu_core_clock(BASE_CLK);
+ run_mode = get_run_clock(BASE_CLK);
+ turbo_mode = get_turbo_clock(BASE_CLK);
+ mem_clk = get_memory_clock(BASE_CLK);
+ sys_clk = get_sysbus_clock(BASE_CLK);
+
+ if (info) {
+ pr_info("Run Mode clock: %ld.%02ldMHz\n",
+ run_mode / 1000000, (run_mode % 1000000) / 10000);
+ pr_info("Turbo Mode clock: %ld.%02ldMHz\n",
+ turbo_mode / 1000000, (turbo_mode % 1000000) / 10000);
+ pr_info("Memory clock: %ld.%02ldMHz\n",
+ mem_clk / 1000000, (mem_clk % 1000000) / 10000);
+ pr_info("System bus clock: %ld.%02ldMHz\n",
+ sys_clk / 1000000, (sys_clk % 1000000) / 10000);
+ }
+
+ return core_clk;
+}
+
+/*
+ * Return the current mem clock frequency as reflected by CCCR[A], B, and L
+ */
+static unsigned long clk_pxa27x_mem_getrate(struct clk_hw *hw,
+ unsigned long parent_rate)
+{
+ return get_memory_clock(BASE_CLK);
+}
+
+static const struct clk_ops clk_pxa27x_mem_ops = {
+ .recalc_rate = clk_pxa27x_mem_getrate,
+};
+
+static unsigned long clk_pxa27x_lcd_getrate(struct clk_hw *hw,
+ unsigned long parent_rate)
+{
+ return get_lcd_clock(BASE_CLK);
+}
+
+static const struct clk_ops clk_pxa27x_lcd_ops = {
+ .recalc_rate = clk_pxa27x_lcd_getrate,
+};
+
+static struct pxa27x_clocks_fixed_cken pxa27x_without_dt_clocks[] __initdata = {
+ PXA2_FIXED_RATE("pxa2xx-uart.0", NULL, FFUART, 14857000, 1),
+ PXA2_FIXED_RATE("pxa2xx-uart.1", NULL, BTUART, 14857000, 1),
+ PXA2_FIXED_RATE("pxa2xx-uart.2", NULL, STUART, 14857000, 1),
+ PXA2_FIXED_RATE(NULL, "UARTCLK", STUART, 14857000, 1),
+ PXA2_FIXED_RATE("pxa2xx-i2s", NULL, I2S, 14682000, 0),
+ PXA2_FIXED_RATE("pxa2xx-i2c.0", NULL, I2C, 32842000, 0),
+ PXA2_FIXED_RATE("pxa27x-udc", NULL, USB, 48000000, 5),
+ PXA2_FIXED_RATE("pxa2xx-mci.0", NULL, MMC, 19500000, 0),
+ PXA2_FIXED_RATE("pxa2xx-ir", "FICPCLK", FICP, 48000000, 0),
+ PXA2_FIXED_RATE("pxa27x-ohci", NULL, USBHOST, 48000000, 0),
+ PXA2_FIXED_RATE("pxa2xx-i2c.1", NULL, PWRI2C, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-keypad", NULL, KEYPAD, 32768, 0),
+ PXA2_FIXED_RATE("pxa27x-ssp.0", NULL, SSP1, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-ssp.1", NULL, SSP2, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-ssp.2", NULL, SSP3, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-pwm.0", NULL, PWM0, 13000000, 0),
+ PXA2_FIXED_RATE("pxa27x-pwm.1", NULL, PWM1, 13000000, 0),
+ PXA2_FIXED_RATE(NULL, "AC97CLK", AC97, 24576000, 0),
+ PXA2_FIXED_RATE(NULL, "AC97CONFCLK", AC97CONF, 24576000, 0),
+ PXA2_FIXED_RATE(NULL, "MSLCLK", MSL, 48000000, 0),
+ PXA2_FIXED_RATE(NULL, "USIMCLK", USIM, 48000000, 0),
+ PXA2_FIXED_RATE(NULL, "MSTKCLK", MEMSTK, 19500000, 0),
+ PXA2_FIXED_RATE(NULL, "IMCLK", IM, 0, 0),
+ PXA2_FIXED_RATE_AO("pxa27x-memc", "MEMCLK", MEMC, 0, 0),
+};
+
+static struct pxa27x_clocks_var_rate pxa27x_var_rate_clocks[] __initdata = {
+ PXA2_VAR_RATE("pxa2xx-fb", true, LCD, &clk_pxa27x_lcd_ops),
+ PXA2_VAR_RATE("pxa27x-camera.0", true, CAMERA, &clk_pxa27x_lcd_ops),
+ PXA2_VAR_RATE("pxa2xx-pcmcia", false, MEMC, &clk_pxa27x_mem_ops),
+};
+
+static int __init pxa27x_clocks_init(void)
+{
+ int i;
+ unsigned long rate, flags;
+ struct clk *clk_parent, *clk;
+ struct pxa27x_clocks_fixed_cken *fclock;
+ struct pxa27x_clocks_var_rate *vclock;
+ char parent_name[80], *name;
+
+ /*
+ * Once device-tree conversion is complete :
+ * if (of_have_populated_dt())
+ * return 0;
+ */
+ for (i = 0; i < ARRAY_SIZE(pxa27x_without_dt_clocks); i++) {
+ fclock = &pxa27x_without_dt_clocks[i];
+ clk_parent = NULL;
+ rate = fclock->rate;
+ snprintf(parent_name, sizeof(parent_name), "clk-%lu", rate);
+ flags = CLK_GET_RATE_NOCACHE | CLK_IS_ROOT;
+ clk_parent = clk_register_fixed_rate(NULL, parent_name, NULL,
+ flags, rate);
+ name = fclock->name ? fclock->name : fclock->con_id;
+ flags = fclock->always_on ? CLK_IGNORE_UNUSED : 0;
+ clk = clk_register_gate(NULL, name, parent_name, flags,
+ (void *)&CKEN, fclock->cken, 0,
+ &cken_lock);
+ if (!IS_ERR(clk))
+ clk_register_clkdev(clk, fclock->con_id, fclock->name);
+ }
+
+ for (i = 0; i < ARRAY_SIZE(pxa27x_var_rate_clocks); i++) {
+ vclock = &pxa27x_var_rate_clocks[i];
+ flags = CLK_GET_RATE_NOCACHE;
+ if (vclock->has_cken) {
+ snprintf(parent_name, sizeof(parent_name), "clk-%s",
+ vclock->name);
+ clk_parent = clk_pxa2xx_register_dummy(parent_name,
+ vclock->clk_ops);
+ clk = clk_register_gate(NULL, vclock->name, parent_name,
+ flags, (void *)&CKEN,
+ vclock->cken, 0, &cken_lock);
+ } else {
+ clk = clk_pxa2xx_register_dummy(vclock->name,
+ vclock->clk_ops);
+ }
+ if (!IS_ERR(clk))
+ clk_register_clkdev(clk, NULL, vclock->name);
+ }
+ return 0;
+}
+
+postcore_initcall(pxa27x_clocks_init);
+
+struct clk_pxa27x_core {
+ struct clk_hw hw;
+ unsigned long (*get_rate)(unsigned long parent_rate);
+};
+
+static unsigned long clk_pxa27x_ops_get_rate(struct clk_hw *hw,
+ unsigned long parent_rate)
+{
+ struct clk_pxa27x_core *clk_core =
+ container_of(hw, struct clk_pxa27x_core, hw);
+
+ return clk_core->get_rate(parent_rate);
+}
+
+const struct clk_ops clk_pxa27x_ops = {
+ .recalc_rate = clk_pxa27x_ops_get_rate,
+};
+
+static struct clk __init *clk_register_pxa27x_core(const char *name,
+ const char *parent_name,
+ unsigned long (*get_rate)(unsigned long))
+{
+ struct clk_pxa27x_core *clk_core;
+ struct clk_init_data init;
+ struct clk *clk;
+
+ clk_core = kzalloc(sizeof(*clk_core), GFP_KERNEL);
+ if (!clk_core) {
+ pr_err("%s: could not allocate pxa27x_core clock\n", __func__);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ init.name = name;
+ init.ops = &clk_pxa27x_ops;
+ init.flags = CLK_IS_BASIC | CLK_GET_RATE_NOCACHE;
+ init.parent_names = (parent_name ? &parent_name : NULL);
+ init.num_parents = (parent_name ? 1 : 0);
+
+ clk_core->get_rate = get_rate;
+ clk_core->hw.init = &init;
+ clk = clk_register(NULL, &clk_core->hw);
+ return clk;
+}
+
+#define NB_CORE_CLOCKS 7
+static struct clk *core_clks[NB_CORE_CLOCKS];
+static struct clk_onecell_data onecell_data = {
+ .clks = core_clks,
+ .clk_num = NB_CORE_CLOCKS,
+};
+
+static void __init pxa27x_core_clocks_init(struct device_node *np)
+{
+ const char *pname;
+
+ pname = of_clk_get_parent_name(np, 0);
+ onecell_data.clks[0] =
+ clk_register_pxa27x_core("run mode", pname, get_run_clock);
+ onecell_data.clks[1] =
+ clk_register_pxa27x_core("turbo mode", pname, get_turbo_clock);
+ onecell_data.clks[2] =
+ clk_register_pxa27x_core("halt-turbo mode", pname,
+ get_halfturbo_clock);
+ onecell_data.clks[3] =
+ clk_register_pxa27x_core("cpu core", pname,
+ get_cpu_core_clock);
+ onecell_data.clks[4] =
+ clk_register_pxa27x_core("system bus", pname,
+ get_sysbus_clock);
+ onecell_data.clks[5] =
+ clk_register_pxa27x_core("memory", pname,
+ get_memory_clock);
+ onecell_data.clks[5] =
+ clk_register_pxa27x_core("lcd", pname,
+ get_lcd_clock);
+
+ of_clk_add_provider(np, of_clk_src_onecell_get, &onecell_data);
+}
+CLK_OF_DECLARE(pxa2xx_clks, "marvell,pxa270-core-clocks",
+ pxa27x_core_clocks_init);
diff --git a/drivers/clk/pxa/clk-pxa2xx.c b/drivers/clk/pxa/clk-pxa2xx.c
new file mode 100644
index 0000000..6e6d657
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa2xx.c
@@ -0,0 +1,74 @@
+/*
+ * Marvell PXA2xx family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/syscore_ops.h>
+#include <mach/pxa2xx-regs.h>
+
+struct dummy_clk {
+ struct clk_hw hw;
+};
+
+struct clk * __init clk_pxa2xx_register_dummy(const char *name,
+ const struct clk_ops *ops)
+{
+ struct clk_init_data init;
+ struct dummy_clk *clock;
+ struct clk *clk;
+
+ clock = kzalloc(sizeof(*clock), GFP_KERNEL);
+ if (!clock) {
+ pr_err("%s: failed to allocate PXA2xx clock.\n", __func__);
+ return ERR_PTR(-ENOMEM);
+ }
+ init.name = name;
+ init.ops = ops;
+ init.parent_names = NULL;
+ init.num_parents = 0;
+ init.flags = CLK_IS_BASIC | CLK_GET_RATE_NOCACHE;
+ clock->hw.init = &init;
+ clk = clk_register(NULL, &clock->hw);
+ if (IS_ERR(clk))
+ kfree(clock);
+ return clk;
+}
+
+
+#ifdef CONFIG_PM
+static uint32_t saved_cken;
+
+static int pxa2xx_clock_suspend(void)
+{
+ saved_cken = CKEN;
+ return 0;
+}
+
+static void pxa2xx_clock_resume(void)
+{
+ CKEN = saved_cken;
+}
+#else
+#define pxa2xx_clock_suspend NULL
+#define pxa2xx_clock_resume NULL
+#endif
+
+static struct syscore_ops pxa2xx_clock_syscore_ops = {
+ .suspend = pxa2xx_clock_suspend,
+ .resume = pxa2xx_clock_resume,
+};
+
+static int __init pxa2xx_clocks_init(void)
+{
+ register_syscore_ops(&pxa2xx_clock_syscore_ops);
+ return 0;
+}
+
+postcore_initcall(pxa2xx_clocks_init);
diff --git a/drivers/clk/pxa/clk-pxa2xx.h b/drivers/clk/pxa/clk-pxa2xx.h
new file mode 100644
index 0000000..79f62af
--- /dev/null
+++ b/drivers/clk/pxa/clk-pxa2xx.h
@@ -0,0 +1,47 @@
+/*
+ * Marvell PXA27x family clocks
+ *
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * For non-devicetree platforms. Once pxa is fully converted to devicetree, this
+ * should go away.
+ */
+
+#ifndef _CLK_PXA2XX_H_
+#define _CLK_PXA2XX_H_
+
+#define PXA2_VAR_RATE(_name, _has_cken, _cken, _clk_ops) \
+ { .name = _name, .has_cken = _has_cken, .cken = CKEN_ ## _cken, \
+ .clk_ops = _clk_ops }
+
+struct pxa27x_clocks_var_rate {
+ char *name;
+ bool has_cken;
+ unsigned int cken;
+ const struct clk_ops *clk_ops;
+};
+
+#define PXA2_FIXED_RATE(_name, _con_id, _cken, _rate, _delay) \
+ { .name = _name, .con_id = _con_id, .cken = CKEN_ ## _cken, \
+ .rate = _rate, .delay = _delay, .always_on = false, }
+#define PXA2_FIXED_RATE_AO(_name, _con_id, _cken, _rate, _delay) \
+ { .name = _name, .con_id = _con_id, .cken = CKEN_ ## _cken, \
+ .rate = _rate, .delay = _delay, .always_on = true, }
+
+struct pxa27x_clocks_fixed_cken {
+ char *name;
+ char *con_id;
+ unsigned int cken;
+ unsigned long rate;
+ int delay;
+ bool always_on;
+};
+
+extern struct clk * __init clk_pxa2xx_register_dummy(const char *name,
+ const struct clk_ops *ops);
+
+#endif
--
2.0.0.rc2
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
2014-06-29 18:32 ` Robert Jarzmik
@ 2014-06-29 18:32 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Haojian Zhuang, Eric Miao
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland,
Robert Jarzmik
Add the clock tree description for the PXA27x based boards.
Signed-off-by: Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>
---
arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++++++++++++++++++++++++-
include/dt-bindings/clock/pxa2xx-clock.h | 45 +++++++++++
2 files changed, 178 insertions(+), 1 deletion(-)
create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
diff --git a/arch/arm/boot/dts/pxa27x.dtsi b/arch/arm/boot/dts/pxa27x.dtsi
index a705469..badaa71 100644
--- a/arch/arm/boot/dts/pxa27x.dtsi
+++ b/arch/arm/boot/dts/pxa27x.dtsi
@@ -1,5 +1,6 @@
/* The pxa3xx skeleton simply augments the 2xx version */
-/include/ "pxa2xx.dtsi"
+#include "pxa2xx.dtsi"
+#include "dt-bindings/clock/pxa2xx-clock.h"
/ {
model = "Marvell PXA27x familiy SoC";
@@ -35,4 +36,135 @@
#pwm-cells = <1>;
};
};
+
+ clocks {
+ /*
+ * The muxing of external clocks/internal dividers for osc* clock
+ * sources has been hidden under the carpet by now.
+ */
+ #address-cells = <1>;
+ #size-cells = <1>;
+ ranges;
+
+ osc13mhz:osc13mhz {
+ compatible = "fixed-clock";
+ #clock-cells = <0>;
+ clock-frequency = <13000000>;
+ };
+
+ osc32_768khz:osc32_768khz {
+ compatible = "fixed-clock";
+ #clock-cells = <0>;
+ clock-frequency = <32768>;
+ };
+
+ pll_312mhz:pll_312mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&osc13mhz>;
+ clock-div = <1>;
+ clock-mult = <24>;
+ };
+
+ clk_48mhz:clk_48mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <13>;
+ clock-mult = <2>;
+ };
+
+ clk_32_842mhz:clk_32_842mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <19>;
+ clock-mult = <2>;
+ };
+
+ clk_19_5mhz:clk_19_5mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <32>;
+ clock-mult = <2>;
+ };
+
+ clk_14_857mhz:clk_14_857mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <42>;
+ clock-mult = <2>;
+ };
+
+ clk_14_682mhz:clk_14_682mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <51>;
+ clock-mult = <2>;
+ };
+
+ clk_13mhz:clk_13mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&osc13mhz>;
+ clock-div = <1>;
+ clock-mult = <1>;
+ };
+
+ clk_dummy:clk_dummy {
+ compatible = "fixed-clock";
+ #clock-cells = <0>;
+ clock-frequency = <0>;
+ };
+
+ clk_ostimer:clk_ostimer {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&osc13mhz>;
+ clock-div = <4>;
+ clock-mult = <1>;
+ };
+
+ pxa27x_sysclks:pxa27x_sysclks {
+ compatible = "marvell,pxa270-core-clocks";
+ #clock-cells = <1>;
+ clocks = <&osc13mhz>;
+ clock-output-names = "run mode", "half-turbo mode",
+ "turbo mode", "cpu core", "system bus", "memory", "lcd";
+ };
+
+ pxa2xx_clks: pxa2xx_clks@41300004 {
+ compatible = "marvell,pxa-clocks";
+ reg = <0x41300004 0x4>;
+ clocks =
+ <&clk_13mhz>, <&clk_13mhz>, <&clk_dummy>, <&clk_13mhz>,
+ <&clk_13mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>,
+ <&clk_14_682mhz>, <&clk_ostimer>, <&clk_48mhz>, <&clk_48mhz>,
+ <&clk_19_5mhz>, <&clk_48mhz>, <&clk_32_842mhz>, <&clk_13mhz>,
+ <&pxa27x_sysclks 6>, <&clk_48mhz>, <&clk_48mhz>, <&osc32_768khz>,
+ <&clk_dummy>, <&clk_19_5mhz>, <&pxa27x_sysclks 4>, <&clk_13mhz>,
+ <&pxa27x_sysclks 6>, <&clk_dummy>;
+ #clock-cells = <1>;
+ clock-output-names =
+ "pwm 0,2", "pwm 1,3", "ac97", "ssp2",
+ "ssp3,hwuart", "stuart", "ffuart", "btuart",
+ "i2s", "nssp,ostimer", "usb host,assp", "usb udc",
+ "mmc", "ficp", "i2c", "pwri2c",
+ "lcd", "msl", "usim", "keypad",
+ "im", "memstk", "memc", "ssp1",
+ "camera", "ac97conf";
+ clock-indices = <
+ CKEN_PWM0 CKEN_PWM1 CKEN_AC97 CKEN_SSP2
+ CKEN_SSP3 CKEN_STUART CKEN_FFUART CKEN_BTUART
+ CKEN_I2S CKEN_OSTIMER CKEN_USBHOST CKEN_USB
+ CKEN_MMC CKEN_FICP CKEN_I2C CKEN_PWRI2C
+ CKEN_LCD CKEN_MSL CKEN_USIM CKEN_KEYPAD
+ CKEN_IM CKEN_MEMSTK 65 CKEN_SSP1
+ CKEN_CAMERA CKEN_AC97CONF >;
+ };
+ };
+
};
diff --git a/include/dt-bindings/clock/pxa2xx-clock.h b/include/dt-bindings/clock/pxa2xx-clock.h
new file mode 100644
index 0000000..5ffba58
--- /dev/null
+++ b/include/dt-bindings/clock/pxa2xx-clock.h
@@ -0,0 +1,45 @@
+/*
+ * Inspired by original work from pxa2xx-regs.h by Nicolas Pitre
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __DT_BINDINGS_CLOCK_PXA2XX_H__
+#define __DT_BINDINGS_CLOCK_PXA2XX_H__
+
+#define CKEN_AC97CONF (31) /* AC97 Controller Configuration */
+#define CKEN_CAMERA (24) /* Camera Interface Clock Enable */
+#define CKEN_SSP1 (23) /* SSP1 Unit Clock Enable */
+#define CKEN_MEMC (22) /* Memory Controller Clock Enable */
+#define CKEN_MEMSTK (21) /* Memory Stick Host Controller */
+#define CKEN_IM (20) /* Internal Memory Clock Enable */
+#define CKEN_KEYPAD (19) /* Keypad Interface Clock Enable */
+#define CKEN_USIM (18) /* USIM Unit Clock Enable */
+#define CKEN_MSL (17) /* MSL Unit Clock Enable */
+#define CKEN_LCD (16) /* LCD Unit Clock Enable */
+#define CKEN_PWRI2C (15) /* PWR I2C Unit Clock Enable */
+#define CKEN_I2C (14) /* I2C Unit Clock Enable */
+#define CKEN_FICP (13) /* FICP Unit Clock Enable */
+#define CKEN_MMC (12) /* MMC Unit Clock Enable */
+#define CKEN_USB (11) /* USB Unit Clock Enable */
+#define CKEN_ASSP (10) /* ASSP (SSP3) Clock Enable */
+#define CKEN_USBHOST (10) /* USB Host Unit Clock Enable */
+#define CKEN_OSTIMER (9) /* OS Timer Unit Clock Enable */
+#define CKEN_NSSP (9) /* NSSP (SSP2) Clock Enable */
+#define CKEN_I2S (8) /* I2S Unit Clock Enable */
+#define CKEN_BTUART (7) /* BTUART Unit Clock Enable */
+#define CKEN_FFUART (6) /* FFUART Unit Clock Enable */
+#define CKEN_STUART (5) /* STUART Unit Clock Enable */
+#define CKEN_HWUART (4) /* HWUART Unit Clock Enable */
+#define CKEN_SSP3 (4) /* SSP3 Unit Clock Enable */
+#define CKEN_SSP (3) /* SSP Unit Clock Enable */
+#define CKEN_SSP2 (3) /* SSP2 Unit Clock Enable */
+#define CKEN_AC97 (2) /* AC97 Unit Clock Enable */
+#define CKEN_PWM1 (1) /* PWM1 Clock Enable */
+#define CKEN_PWM0 (0) /* PWM0 Clock Enable */
+
+#endif
--
2.0.0.rc2
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
@ 2014-06-29 18:32 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: linux-arm-kernel
Add the clock tree description for the PXA27x based boards.
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
---
arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++++++++++++++++++++++++-
include/dt-bindings/clock/pxa2xx-clock.h | 45 +++++++++++
2 files changed, 178 insertions(+), 1 deletion(-)
create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
diff --git a/arch/arm/boot/dts/pxa27x.dtsi b/arch/arm/boot/dts/pxa27x.dtsi
index a705469..badaa71 100644
--- a/arch/arm/boot/dts/pxa27x.dtsi
+++ b/arch/arm/boot/dts/pxa27x.dtsi
@@ -1,5 +1,6 @@
/* The pxa3xx skeleton simply augments the 2xx version */
-/include/ "pxa2xx.dtsi"
+#include "pxa2xx.dtsi"
+#include "dt-bindings/clock/pxa2xx-clock.h"
/ {
model = "Marvell PXA27x familiy SoC";
@@ -35,4 +36,135 @@
#pwm-cells = <1>;
};
};
+
+ clocks {
+ /*
+ * The muxing of external clocks/internal dividers for osc* clock
+ * sources has been hidden under the carpet by now.
+ */
+ #address-cells = <1>;
+ #size-cells = <1>;
+ ranges;
+
+ osc13mhz:osc13mhz {
+ compatible = "fixed-clock";
+ #clock-cells = <0>;
+ clock-frequency = <13000000>;
+ };
+
+ osc32_768khz:osc32_768khz {
+ compatible = "fixed-clock";
+ #clock-cells = <0>;
+ clock-frequency = <32768>;
+ };
+
+ pll_312mhz:pll_312mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&osc13mhz>;
+ clock-div = <1>;
+ clock-mult = <24>;
+ };
+
+ clk_48mhz:clk_48mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <13>;
+ clock-mult = <2>;
+ };
+
+ clk_32_842mhz:clk_32_842mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <19>;
+ clock-mult = <2>;
+ };
+
+ clk_19_5mhz:clk_19_5mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <32>;
+ clock-mult = <2>;
+ };
+
+ clk_14_857mhz:clk_14_857mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <42>;
+ clock-mult = <2>;
+ };
+
+ clk_14_682mhz:clk_14_682mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&pll_312mhz>;
+ clock-div = <51>;
+ clock-mult = <2>;
+ };
+
+ clk_13mhz:clk_13mhz {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&osc13mhz>;
+ clock-div = <1>;
+ clock-mult = <1>;
+ };
+
+ clk_dummy:clk_dummy {
+ compatible = "fixed-clock";
+ #clock-cells = <0>;
+ clock-frequency = <0>;
+ };
+
+ clk_ostimer:clk_ostimer {
+ compatible = "fixed-factor-clock";
+ #clock-cells = <0>;
+ clocks = <&osc13mhz>;
+ clock-div = <4>;
+ clock-mult = <1>;
+ };
+
+ pxa27x_sysclks:pxa27x_sysclks {
+ compatible = "marvell,pxa270-core-clocks";
+ #clock-cells = <1>;
+ clocks = <&osc13mhz>;
+ clock-output-names = "run mode", "half-turbo mode",
+ "turbo mode", "cpu core", "system bus", "memory", "lcd";
+ };
+
+ pxa2xx_clks: pxa2xx_clks at 41300004 {
+ compatible = "marvell,pxa-clocks";
+ reg = <0x41300004 0x4>;
+ clocks =
+ <&clk_13mhz>, <&clk_13mhz>, <&clk_dummy>, <&clk_13mhz>,
+ <&clk_13mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>,
+ <&clk_14_682mhz>, <&clk_ostimer>, <&clk_48mhz>, <&clk_48mhz>,
+ <&clk_19_5mhz>, <&clk_48mhz>, <&clk_32_842mhz>, <&clk_13mhz>,
+ <&pxa27x_sysclks 6>, <&clk_48mhz>, <&clk_48mhz>, <&osc32_768khz>,
+ <&clk_dummy>, <&clk_19_5mhz>, <&pxa27x_sysclks 4>, <&clk_13mhz>,
+ <&pxa27x_sysclks 6>, <&clk_dummy>;
+ #clock-cells = <1>;
+ clock-output-names =
+ "pwm 0,2", "pwm 1,3", "ac97", "ssp2",
+ "ssp3,hwuart", "stuart", "ffuart", "btuart",
+ "i2s", "nssp,ostimer", "usb host,assp", "usb udc",
+ "mmc", "ficp", "i2c", "pwri2c",
+ "lcd", "msl", "usim", "keypad",
+ "im", "memstk", "memc", "ssp1",
+ "camera", "ac97conf";
+ clock-indices = <
+ CKEN_PWM0 CKEN_PWM1 CKEN_AC97 CKEN_SSP2
+ CKEN_SSP3 CKEN_STUART CKEN_FFUART CKEN_BTUART
+ CKEN_I2S CKEN_OSTIMER CKEN_USBHOST CKEN_USB
+ CKEN_MMC CKEN_FICP CKEN_I2C CKEN_PWRI2C
+ CKEN_LCD CKEN_MSL CKEN_USIM CKEN_KEYPAD
+ CKEN_IM CKEN_MEMSTK 65 CKEN_SSP1
+ CKEN_CAMERA CKEN_AC97CONF >;
+ };
+ };
+
};
diff --git a/include/dt-bindings/clock/pxa2xx-clock.h b/include/dt-bindings/clock/pxa2xx-clock.h
new file mode 100644
index 0000000..5ffba58
--- /dev/null
+++ b/include/dt-bindings/clock/pxa2xx-clock.h
@@ -0,0 +1,45 @@
+/*
+ * Inspired by original work from pxa2xx-regs.h by Nicolas Pitre
+ * Copyright (C) 2014 Robert Jarzmik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __DT_BINDINGS_CLOCK_PXA2XX_H__
+#define __DT_BINDINGS_CLOCK_PXA2XX_H__
+
+#define CKEN_AC97CONF (31) /* AC97 Controller Configuration */
+#define CKEN_CAMERA (24) /* Camera Interface Clock Enable */
+#define CKEN_SSP1 (23) /* SSP1 Unit Clock Enable */
+#define CKEN_MEMC (22) /* Memory Controller Clock Enable */
+#define CKEN_MEMSTK (21) /* Memory Stick Host Controller */
+#define CKEN_IM (20) /* Internal Memory Clock Enable */
+#define CKEN_KEYPAD (19) /* Keypad Interface Clock Enable */
+#define CKEN_USIM (18) /* USIM Unit Clock Enable */
+#define CKEN_MSL (17) /* MSL Unit Clock Enable */
+#define CKEN_LCD (16) /* LCD Unit Clock Enable */
+#define CKEN_PWRI2C (15) /* PWR I2C Unit Clock Enable */
+#define CKEN_I2C (14) /* I2C Unit Clock Enable */
+#define CKEN_FICP (13) /* FICP Unit Clock Enable */
+#define CKEN_MMC (12) /* MMC Unit Clock Enable */
+#define CKEN_USB (11) /* USB Unit Clock Enable */
+#define CKEN_ASSP (10) /* ASSP (SSP3) Clock Enable */
+#define CKEN_USBHOST (10) /* USB Host Unit Clock Enable */
+#define CKEN_OSTIMER (9) /* OS Timer Unit Clock Enable */
+#define CKEN_NSSP (9) /* NSSP (SSP2) Clock Enable */
+#define CKEN_I2S (8) /* I2S Unit Clock Enable */
+#define CKEN_BTUART (7) /* BTUART Unit Clock Enable */
+#define CKEN_FFUART (6) /* FFUART Unit Clock Enable */
+#define CKEN_STUART (5) /* STUART Unit Clock Enable */
+#define CKEN_HWUART (4) /* HWUART Unit Clock Enable */
+#define CKEN_SSP3 (4) /* SSP3 Unit Clock Enable */
+#define CKEN_SSP (3) /* SSP Unit Clock Enable */
+#define CKEN_SSP2 (3) /* SSP2 Unit Clock Enable */
+#define CKEN_AC97 (2) /* AC97 Unit Clock Enable */
+#define CKEN_PWM1 (1) /* PWM1 Clock Enable */
+#define CKEN_PWM0 (0) /* PWM0 Clock Enable */
+
+#endif
--
2.0.0.rc2
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [PATCH 3/4] arm: pxa: Transition pxa27x to clk framework
2014-06-29 18:32 ` Robert Jarzmik
@ 2014-06-29 18:32 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Haojian Zhuang, Eric Miao
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland,
Robert Jarzmik
Transition the PXA27x CPUs to the clock framework.
This transition still enables legacy platforms to run without device
tree as before, ie relying on platform data encoded in board specific
files.
Signed-off-by: Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>
---
arch/arm/Kconfig | 1 +
arch/arm/mach-pxa/Makefile | 8 +-
arch/arm/mach-pxa/pxa27x.c | 190 ++++-----------------------------------------
3 files changed, 21 insertions(+), 178 deletions(-)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 87b63fd..96bdc7b 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -642,6 +642,7 @@ config ARCH_PXA
select ARCH_REQUIRE_GPIOLIB
select ARM_CPU_SUSPEND if PM
select AUTO_ZRELADDR
+ select COMMON_CLK if PXA27x
select CLKDEV_LOOKUP
select CLKSRC_MMIO
select GENERIC_CLOCKEVENTS
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index 648867a..ce15823 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -3,16 +3,16 @@
#
# Common support (must be linked before board specific support)
-obj-y += clock.o devices.o generic.o irq.o \
+obj-y += devices.o generic.o irq.o \
time.o reset.o
obj-$(CONFIG_PM) += pm.o sleep.o standby.o
# Generic drivers that other drivers may depend upon
# SoC-specific code
-obj-$(CONFIG_PXA25x) += mfp-pxa2xx.o clock-pxa2xx.o pxa2xx.o pxa25x.o
-obj-$(CONFIG_PXA27x) += mfp-pxa2xx.o clock-pxa2xx.o pxa2xx.o pxa27x.o
-obj-$(CONFIG_PXA3xx) += mfp-pxa3xx.o clock-pxa3xx.o pxa3xx.o smemc.o pxa3xx-ulpi.o
+obj-$(CONFIG_PXA25x) += mfp-pxa2xx.o clock.o clock-pxa2xx.o pxa2xx.o pxa25x.o
+obj-$(CONFIG_PXA27x) += mfp-pxa2xx.o pxa2xx.o pxa27x.o
+obj-$(CONFIG_PXA3xx) += mfp-pxa3xx.o clock.o clock-pxa3xx.o pxa3xx.o smemc.o pxa3xx-ulpi.o
obj-$(CONFIG_CPU_PXA300) += pxa300.o
obj-$(CONFIG_CPU_PXA320) += pxa320.o
obj-$(CONFIG_CPU_PXA930) += pxa930.o
diff --git a/arch/arm/mach-pxa/pxa27x.c b/arch/arm/mach-pxa/pxa27x.c
index 301471a..a8e229fd 100644
--- a/arch/arm/mach-pxa/pxa27x.c
+++ b/arch/arm/mach-pxa/pxa27x.c
@@ -37,7 +37,8 @@
#include "generic.h"
#include "devices.h"
-#include "clock.h"
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
void pxa27x_clear_otgph(void)
{
@@ -73,174 +74,6 @@ void pxa27x_configure_ac97reset(int reset_gpio, bool to_gpio)
}
EXPORT_SYMBOL_GPL(pxa27x_configure_ac97reset);
-/* Crystal clock: 13MHz */
-#define BASE_CLK 13000000
-
-/*
- * Get the clock frequency as reflected by CCSR and the turbo flag.
- * We assume these values have been applied via a fcs.
- * If info is not 0 we also display the current settings.
- */
-unsigned int pxa27x_get_clk_frequency_khz(int info)
-{
- unsigned long ccsr, clkcfg;
- unsigned int l, L, m, M, n2, N, S;
- int cccr_a, t, ht, b;
-
- ccsr = CCSR;
- cccr_a = CCCR & (1 << 25);
-
- /* Read clkcfg register: it has turbo, b, half-turbo (and f) */
- asm( "mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg) );
- t = clkcfg & (1 << 0);
- ht = clkcfg & (1 << 2);
- b = clkcfg & (1 << 3);
-
- l = ccsr & 0x1f;
- n2 = (ccsr>>7) & 0xf;
- m = (l <= 10) ? 1 : (l <= 20) ? 2 : 4;
-
- L = l * BASE_CLK;
- N = (L * n2) / 2;
- M = (!cccr_a) ? (L/m) : ((b) ? L : (L/2));
- S = (b) ? L : (L/2);
-
- if (info) {
- printk( KERN_INFO "Run Mode clock: %d.%02dMHz (*%d)\n",
- L / 1000000, (L % 1000000) / 10000, l );
- printk( KERN_INFO "Turbo Mode clock: %d.%02dMHz (*%d.%d, %sactive)\n",
- N / 1000000, (N % 1000000)/10000, n2 / 2, (n2 % 2)*5,
- (t) ? "" : "in" );
- printk( KERN_INFO "Memory clock: %d.%02dMHz (/%d)\n",
- M / 1000000, (M % 1000000) / 10000, m );
- printk( KERN_INFO "System bus clock: %d.%02dMHz \n",
- S / 1000000, (S % 1000000) / 10000 );
- }
-
- return (t) ? (N/1000) : (L/1000);
-}
-
-/*
- * Return the current mem clock frequency as reflected by CCCR[A], B, and L
- */
-static unsigned long clk_pxa27x_mem_getrate(struct clk *clk)
-{
- unsigned long ccsr, clkcfg;
- unsigned int l, L, m, M;
- int cccr_a, b;
-
- ccsr = CCSR;
- cccr_a = CCCR & (1 << 25);
-
- /* Read clkcfg register: it has turbo, b, half-turbo (and f) */
- asm( "mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg) );
- b = clkcfg & (1 << 3);
-
- l = ccsr & 0x1f;
- m = (l <= 10) ? 1 : (l <= 20) ? 2 : 4;
-
- L = l * BASE_CLK;
- M = (!cccr_a) ? (L/m) : ((b) ? L : (L/2));
-
- return M;
-}
-
-static const struct clkops clk_pxa27x_mem_ops = {
- .enable = clk_dummy_enable,
- .disable = clk_dummy_disable,
- .getrate = clk_pxa27x_mem_getrate,
-};
-
-/*
- * Return the current LCD clock frequency in units of 10kHz as
- */
-static unsigned int pxa27x_get_lcdclk_frequency_10khz(void)
-{
- unsigned long ccsr;
- unsigned int l, L, k, K;
-
- ccsr = CCSR;
-
- l = ccsr & 0x1f;
- k = (l <= 7) ? 1 : (l <= 16) ? 2 : 4;
-
- L = l * BASE_CLK;
- K = L / k;
-
- return (K / 10000);
-}
-
-static unsigned long clk_pxa27x_lcd_getrate(struct clk *clk)
-{
- return pxa27x_get_lcdclk_frequency_10khz() * 10000;
-}
-
-static const struct clkops clk_pxa27x_lcd_ops = {
- .enable = clk_pxa2xx_cken_enable,
- .disable = clk_pxa2xx_cken_disable,
- .getrate = clk_pxa27x_lcd_getrate,
-};
-
-static DEFINE_PXA2_CKEN(pxa27x_ffuart, FFUART, 14857000, 1);
-static DEFINE_PXA2_CKEN(pxa27x_btuart, BTUART, 14857000, 1);
-static DEFINE_PXA2_CKEN(pxa27x_stuart, STUART, 14857000, 1);
-static DEFINE_PXA2_CKEN(pxa27x_i2s, I2S, 14682000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_i2c, I2C, 32842000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_usb, USB, 48000000, 5);
-static DEFINE_PXA2_CKEN(pxa27x_mmc, MMC, 19500000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ficp, FICP, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_usbhost, USBHOST, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_pwri2c, PWRI2C, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_keypad, KEYPAD, 32768, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ssp1, SSP1, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ssp2, SSP2, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ssp3, SSP3, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_pwm0, PWM0, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_pwm1, PWM1, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ac97, AC97, 24576000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ac97conf, AC97CONF, 24576000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_msl, MSL, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_usim, USIM, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_memstk, MEMSTK, 19500000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_im, IM, 0, 0);
-static DEFINE_PXA2_CKEN(pxa27x_memc, MEMC, 0, 0);
-
-static DEFINE_CK(pxa27x_lcd, LCD, &clk_pxa27x_lcd_ops);
-static DEFINE_CK(pxa27x_camera, CAMERA, &clk_pxa27x_lcd_ops);
-static DEFINE_CLK(pxa27x_mem, &clk_pxa27x_mem_ops, 0, 0);
-
-static struct clk_lookup pxa27x_clkregs[] = {
- INIT_CLKREG(&clk_pxa27x_lcd, "pxa2xx-fb", NULL),
- INIT_CLKREG(&clk_pxa27x_camera, "pxa27x-camera.0", NULL),
- INIT_CLKREG(&clk_pxa27x_ffuart, "pxa2xx-uart.0", NULL),
- INIT_CLKREG(&clk_pxa27x_btuart, "pxa2xx-uart.1", NULL),
- INIT_CLKREG(&clk_pxa27x_stuart, "pxa2xx-uart.2", NULL),
- INIT_CLKREG(&clk_pxa27x_i2s, "pxa2xx-i2s", NULL),
- INIT_CLKREG(&clk_pxa27x_i2c, "pxa2xx-i2c.0", NULL),
- INIT_CLKREG(&clk_pxa27x_usb, "pxa27x-udc", NULL),
- INIT_CLKREG(&clk_pxa27x_mmc, "pxa2xx-mci.0", NULL),
- INIT_CLKREG(&clk_pxa27x_stuart, "pxa2xx-ir", "UARTCLK"),
- INIT_CLKREG(&clk_pxa27x_ficp, "pxa2xx-ir", "FICPCLK"),
- INIT_CLKREG(&clk_pxa27x_usbhost, "pxa27x-ohci", NULL),
- INIT_CLKREG(&clk_pxa27x_pwri2c, "pxa2xx-i2c.1", NULL),
- INIT_CLKREG(&clk_pxa27x_keypad, "pxa27x-keypad", NULL),
- INIT_CLKREG(&clk_pxa27x_ssp1, "pxa27x-ssp.0", NULL),
- INIT_CLKREG(&clk_pxa27x_ssp2, "pxa27x-ssp.1", NULL),
- INIT_CLKREG(&clk_pxa27x_ssp3, "pxa27x-ssp.2", NULL),
- INIT_CLKREG(&clk_pxa27x_pwm0, "pxa27x-pwm.0", NULL),
- INIT_CLKREG(&clk_pxa27x_pwm1, "pxa27x-pwm.1", NULL),
- INIT_CLKREG(&clk_pxa27x_ac97, NULL, "AC97CLK"),
- INIT_CLKREG(&clk_pxa27x_ac97conf, NULL, "AC97CONFCLK"),
- INIT_CLKREG(&clk_pxa27x_msl, NULL, "MSLCLK"),
- INIT_CLKREG(&clk_pxa27x_usim, NULL, "USIMCLK"),
- INIT_CLKREG(&clk_pxa27x_memstk, NULL, "MSTKCLK"),
- INIT_CLKREG(&clk_pxa27x_im, NULL, "IMCLK"),
- INIT_CLKREG(&clk_pxa27x_memc, NULL, "MEMCLK"),
- INIT_CLKREG(&clk_pxa27x_mem, "pxa2xx-pcmcia", NULL),
- INIT_CLKREG(&clk_dummy, "pxa27x-gpio", NULL),
- INIT_CLKREG(&clk_dummy, "sa1100-rtc", NULL),
-};
-
#ifdef CONFIG_PM
#define SAVE(x) sleep_save[SLEEP_SAVE_##x] = x
@@ -452,6 +285,8 @@ static struct platform_device *devices[] __initdata = {
&pxa27x_device_pwm1,
};
+static struct clk *clk_gpio;
+
static int __init pxa27x_init(void)
{
int ret = 0;
@@ -460,8 +295,6 @@ static int __init pxa27x_init(void)
reset_status = RCSR;
- clkdev_add_table(pxa27x_clkregs, ARRAY_SIZE(pxa27x_clkregs));
-
if ((ret = pxa_init_dma(IRQ_DMA, 32)))
return ret;
@@ -469,10 +302,19 @@ static int __init pxa27x_init(void)
register_syscore_ops(&pxa_irq_syscore_ops);
register_syscore_ops(&pxa2xx_mfp_syscore_ops);
- register_syscore_ops(&pxa2xx_clock_syscore_ops);
- pxa_register_device(&pxa27x_device_gpio, &pxa27x_gpio_info);
- ret = platform_add_devices(devices, ARRAY_SIZE(devices));
+ if (!of_have_populated_dt()) {
+ clk_gpio = clk_register_fixed_rate(NULL, "pxa27x-gpio",
+ NULL, 0, 1);
+ ret = clk_register_clkdev(clk_gpio, NULL,
+ "pxa27x-gpio");
+ if (!ret)
+ pxa_register_device(&pxa27x_device_gpio,
+ &pxa27x_gpio_info);
+ if (!ret)
+ ret = platform_add_devices(devices,
+ ARRAY_SIZE(devices));
+ }
}
return ret;
--
2.0.0.rc2
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [PATCH 3/4] arm: pxa: Transition pxa27x to clk framework
@ 2014-06-29 18:32 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: linux-arm-kernel
Transition the PXA27x CPUs to the clock framework.
This transition still enables legacy platforms to run without device
tree as before, ie relying on platform data encoded in board specific
files.
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
---
arch/arm/Kconfig | 1 +
arch/arm/mach-pxa/Makefile | 8 +-
arch/arm/mach-pxa/pxa27x.c | 190 ++++-----------------------------------------
3 files changed, 21 insertions(+), 178 deletions(-)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 87b63fd..96bdc7b 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -642,6 +642,7 @@ config ARCH_PXA
select ARCH_REQUIRE_GPIOLIB
select ARM_CPU_SUSPEND if PM
select AUTO_ZRELADDR
+ select COMMON_CLK if PXA27x
select CLKDEV_LOOKUP
select CLKSRC_MMIO
select GENERIC_CLOCKEVENTS
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index 648867a..ce15823 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -3,16 +3,16 @@
#
# Common support (must be linked before board specific support)
-obj-y += clock.o devices.o generic.o irq.o \
+obj-y += devices.o generic.o irq.o \
time.o reset.o
obj-$(CONFIG_PM) += pm.o sleep.o standby.o
# Generic drivers that other drivers may depend upon
# SoC-specific code
-obj-$(CONFIG_PXA25x) += mfp-pxa2xx.o clock-pxa2xx.o pxa2xx.o pxa25x.o
-obj-$(CONFIG_PXA27x) += mfp-pxa2xx.o clock-pxa2xx.o pxa2xx.o pxa27x.o
-obj-$(CONFIG_PXA3xx) += mfp-pxa3xx.o clock-pxa3xx.o pxa3xx.o smemc.o pxa3xx-ulpi.o
+obj-$(CONFIG_PXA25x) += mfp-pxa2xx.o clock.o clock-pxa2xx.o pxa2xx.o pxa25x.o
+obj-$(CONFIG_PXA27x) += mfp-pxa2xx.o pxa2xx.o pxa27x.o
+obj-$(CONFIG_PXA3xx) += mfp-pxa3xx.o clock.o clock-pxa3xx.o pxa3xx.o smemc.o pxa3xx-ulpi.o
obj-$(CONFIG_CPU_PXA300) += pxa300.o
obj-$(CONFIG_CPU_PXA320) += pxa320.o
obj-$(CONFIG_CPU_PXA930) += pxa930.o
diff --git a/arch/arm/mach-pxa/pxa27x.c b/arch/arm/mach-pxa/pxa27x.c
index 301471a..a8e229fd 100644
--- a/arch/arm/mach-pxa/pxa27x.c
+++ b/arch/arm/mach-pxa/pxa27x.c
@@ -37,7 +37,8 @@
#include "generic.h"
#include "devices.h"
-#include "clock.h"
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
void pxa27x_clear_otgph(void)
{
@@ -73,174 +74,6 @@ void pxa27x_configure_ac97reset(int reset_gpio, bool to_gpio)
}
EXPORT_SYMBOL_GPL(pxa27x_configure_ac97reset);
-/* Crystal clock: 13MHz */
-#define BASE_CLK 13000000
-
-/*
- * Get the clock frequency as reflected by CCSR and the turbo flag.
- * We assume these values have been applied via a fcs.
- * If info is not 0 we also display the current settings.
- */
-unsigned int pxa27x_get_clk_frequency_khz(int info)
-{
- unsigned long ccsr, clkcfg;
- unsigned int l, L, m, M, n2, N, S;
- int cccr_a, t, ht, b;
-
- ccsr = CCSR;
- cccr_a = CCCR & (1 << 25);
-
- /* Read clkcfg register: it has turbo, b, half-turbo (and f) */
- asm( "mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg) );
- t = clkcfg & (1 << 0);
- ht = clkcfg & (1 << 2);
- b = clkcfg & (1 << 3);
-
- l = ccsr & 0x1f;
- n2 = (ccsr>>7) & 0xf;
- m = (l <= 10) ? 1 : (l <= 20) ? 2 : 4;
-
- L = l * BASE_CLK;
- N = (L * n2) / 2;
- M = (!cccr_a) ? (L/m) : ((b) ? L : (L/2));
- S = (b) ? L : (L/2);
-
- if (info) {
- printk( KERN_INFO "Run Mode clock: %d.%02dMHz (*%d)\n",
- L / 1000000, (L % 1000000) / 10000, l );
- printk( KERN_INFO "Turbo Mode clock: %d.%02dMHz (*%d.%d, %sactive)\n",
- N / 1000000, (N % 1000000)/10000, n2 / 2, (n2 % 2)*5,
- (t) ? "" : "in" );
- printk( KERN_INFO "Memory clock: %d.%02dMHz (/%d)\n",
- M / 1000000, (M % 1000000) / 10000, m );
- printk( KERN_INFO "System bus clock: %d.%02dMHz \n",
- S / 1000000, (S % 1000000) / 10000 );
- }
-
- return (t) ? (N/1000) : (L/1000);
-}
-
-/*
- * Return the current mem clock frequency as reflected by CCCR[A], B, and L
- */
-static unsigned long clk_pxa27x_mem_getrate(struct clk *clk)
-{
- unsigned long ccsr, clkcfg;
- unsigned int l, L, m, M;
- int cccr_a, b;
-
- ccsr = CCSR;
- cccr_a = CCCR & (1 << 25);
-
- /* Read clkcfg register: it has turbo, b, half-turbo (and f) */
- asm( "mrc\tp14, 0, %0, c6, c0, 0" : "=r" (clkcfg) );
- b = clkcfg & (1 << 3);
-
- l = ccsr & 0x1f;
- m = (l <= 10) ? 1 : (l <= 20) ? 2 : 4;
-
- L = l * BASE_CLK;
- M = (!cccr_a) ? (L/m) : ((b) ? L : (L/2));
-
- return M;
-}
-
-static const struct clkops clk_pxa27x_mem_ops = {
- .enable = clk_dummy_enable,
- .disable = clk_dummy_disable,
- .getrate = clk_pxa27x_mem_getrate,
-};
-
-/*
- * Return the current LCD clock frequency in units of 10kHz as
- */
-static unsigned int pxa27x_get_lcdclk_frequency_10khz(void)
-{
- unsigned long ccsr;
- unsigned int l, L, k, K;
-
- ccsr = CCSR;
-
- l = ccsr & 0x1f;
- k = (l <= 7) ? 1 : (l <= 16) ? 2 : 4;
-
- L = l * BASE_CLK;
- K = L / k;
-
- return (K / 10000);
-}
-
-static unsigned long clk_pxa27x_lcd_getrate(struct clk *clk)
-{
- return pxa27x_get_lcdclk_frequency_10khz() * 10000;
-}
-
-static const struct clkops clk_pxa27x_lcd_ops = {
- .enable = clk_pxa2xx_cken_enable,
- .disable = clk_pxa2xx_cken_disable,
- .getrate = clk_pxa27x_lcd_getrate,
-};
-
-static DEFINE_PXA2_CKEN(pxa27x_ffuart, FFUART, 14857000, 1);
-static DEFINE_PXA2_CKEN(pxa27x_btuart, BTUART, 14857000, 1);
-static DEFINE_PXA2_CKEN(pxa27x_stuart, STUART, 14857000, 1);
-static DEFINE_PXA2_CKEN(pxa27x_i2s, I2S, 14682000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_i2c, I2C, 32842000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_usb, USB, 48000000, 5);
-static DEFINE_PXA2_CKEN(pxa27x_mmc, MMC, 19500000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ficp, FICP, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_usbhost, USBHOST, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_pwri2c, PWRI2C, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_keypad, KEYPAD, 32768, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ssp1, SSP1, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ssp2, SSP2, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ssp3, SSP3, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_pwm0, PWM0, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_pwm1, PWM1, 13000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ac97, AC97, 24576000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_ac97conf, AC97CONF, 24576000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_msl, MSL, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_usim, USIM, 48000000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_memstk, MEMSTK, 19500000, 0);
-static DEFINE_PXA2_CKEN(pxa27x_im, IM, 0, 0);
-static DEFINE_PXA2_CKEN(pxa27x_memc, MEMC, 0, 0);
-
-static DEFINE_CK(pxa27x_lcd, LCD, &clk_pxa27x_lcd_ops);
-static DEFINE_CK(pxa27x_camera, CAMERA, &clk_pxa27x_lcd_ops);
-static DEFINE_CLK(pxa27x_mem, &clk_pxa27x_mem_ops, 0, 0);
-
-static struct clk_lookup pxa27x_clkregs[] = {
- INIT_CLKREG(&clk_pxa27x_lcd, "pxa2xx-fb", NULL),
- INIT_CLKREG(&clk_pxa27x_camera, "pxa27x-camera.0", NULL),
- INIT_CLKREG(&clk_pxa27x_ffuart, "pxa2xx-uart.0", NULL),
- INIT_CLKREG(&clk_pxa27x_btuart, "pxa2xx-uart.1", NULL),
- INIT_CLKREG(&clk_pxa27x_stuart, "pxa2xx-uart.2", NULL),
- INIT_CLKREG(&clk_pxa27x_i2s, "pxa2xx-i2s", NULL),
- INIT_CLKREG(&clk_pxa27x_i2c, "pxa2xx-i2c.0", NULL),
- INIT_CLKREG(&clk_pxa27x_usb, "pxa27x-udc", NULL),
- INIT_CLKREG(&clk_pxa27x_mmc, "pxa2xx-mci.0", NULL),
- INIT_CLKREG(&clk_pxa27x_stuart, "pxa2xx-ir", "UARTCLK"),
- INIT_CLKREG(&clk_pxa27x_ficp, "pxa2xx-ir", "FICPCLK"),
- INIT_CLKREG(&clk_pxa27x_usbhost, "pxa27x-ohci", NULL),
- INIT_CLKREG(&clk_pxa27x_pwri2c, "pxa2xx-i2c.1", NULL),
- INIT_CLKREG(&clk_pxa27x_keypad, "pxa27x-keypad", NULL),
- INIT_CLKREG(&clk_pxa27x_ssp1, "pxa27x-ssp.0", NULL),
- INIT_CLKREG(&clk_pxa27x_ssp2, "pxa27x-ssp.1", NULL),
- INIT_CLKREG(&clk_pxa27x_ssp3, "pxa27x-ssp.2", NULL),
- INIT_CLKREG(&clk_pxa27x_pwm0, "pxa27x-pwm.0", NULL),
- INIT_CLKREG(&clk_pxa27x_pwm1, "pxa27x-pwm.1", NULL),
- INIT_CLKREG(&clk_pxa27x_ac97, NULL, "AC97CLK"),
- INIT_CLKREG(&clk_pxa27x_ac97conf, NULL, "AC97CONFCLK"),
- INIT_CLKREG(&clk_pxa27x_msl, NULL, "MSLCLK"),
- INIT_CLKREG(&clk_pxa27x_usim, NULL, "USIMCLK"),
- INIT_CLKREG(&clk_pxa27x_memstk, NULL, "MSTKCLK"),
- INIT_CLKREG(&clk_pxa27x_im, NULL, "IMCLK"),
- INIT_CLKREG(&clk_pxa27x_memc, NULL, "MEMCLK"),
- INIT_CLKREG(&clk_pxa27x_mem, "pxa2xx-pcmcia", NULL),
- INIT_CLKREG(&clk_dummy, "pxa27x-gpio", NULL),
- INIT_CLKREG(&clk_dummy, "sa1100-rtc", NULL),
-};
-
#ifdef CONFIG_PM
#define SAVE(x) sleep_save[SLEEP_SAVE_##x] = x
@@ -452,6 +285,8 @@ static struct platform_device *devices[] __initdata = {
&pxa27x_device_pwm1,
};
+static struct clk *clk_gpio;
+
static int __init pxa27x_init(void)
{
int ret = 0;
@@ -460,8 +295,6 @@ static int __init pxa27x_init(void)
reset_status = RCSR;
- clkdev_add_table(pxa27x_clkregs, ARRAY_SIZE(pxa27x_clkregs));
-
if ((ret = pxa_init_dma(IRQ_DMA, 32)))
return ret;
@@ -469,10 +302,19 @@ static int __init pxa27x_init(void)
register_syscore_ops(&pxa_irq_syscore_ops);
register_syscore_ops(&pxa2xx_mfp_syscore_ops);
- register_syscore_ops(&pxa2xx_clock_syscore_ops);
- pxa_register_device(&pxa27x_device_gpio, &pxa27x_gpio_info);
- ret = platform_add_devices(devices, ARRAY_SIZE(devices));
+ if (!of_have_populated_dt()) {
+ clk_gpio = clk_register_fixed_rate(NULL, "pxa27x-gpio",
+ NULL, 0, 1);
+ ret = clk_register_clkdev(clk_gpio, NULL,
+ "pxa27x-gpio");
+ if (!ret)
+ pxa_register_device(&pxa27x_device_gpio,
+ &pxa27x_gpio_info);
+ if (!ret)
+ ret = platform_add_devices(devices,
+ ARRAY_SIZE(devices));
+ }
}
return ret;
--
2.0.0.rc2
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [PATCH 4/4] clk: dts: document pxa27x clock binding
2014-06-29 18:32 ` Robert Jarzmik
@ 2014-06-29 18:32 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Haojian Zhuang, Eric Miao
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland,
Robert Jarzmik
Document the device-tree binding of Marvell PXA based SoCs.
PXA clocks are mostly fixed rate and fixed ratio clocks derived from an
external oscillator, and gated by a register set (CKEN or CKEN*).
Signed-off-by: Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>
---
.../devicetree/bindings/clock/pxa-clock.txt | 32 ++++++++++++++++++++++
1 file changed, 32 insertions(+)
create mode 100644 Documentation/devicetree/bindings/clock/pxa-clock.txt
diff --git a/Documentation/devicetree/bindings/clock/pxa-clock.txt b/Documentation/devicetree/bindings/clock/pxa-clock.txt
new file mode 100644
index 0000000..beb28ea
--- /dev/null
+++ b/Documentation/devicetree/bindings/clock/pxa-clock.txt
@@ -0,0 +1,32 @@
+* Clock bindings for Marvell PXA chips
+
+Required properties:
+- compatible: Should be "marvell,pxa-clocks"
+- reg: Address and length of the CKEN register
+- #clock-cells: Should be <1>
+- clocks: the parent fixed rate clocks
+- clock-indices: the bit index in the CKEN register gating the clock
+- clock-outpout-names: the names of the clocks
+
+The clock consumer should specify the desired clock by having the clock
+ID in its "clocks" phandle cell.
+
+Examples:
+
+pxa2xx_clks: pxa2xx_clks@41300004 {
+ compatible = "marvell,pxa-clocks";
+ clocks = <&clk_14_857mhz>, <&clk_14_857mhz>
+ #clock-cells = <1>;
+ clock-output-names =
+ "pxa27x-ffuart", "pxa27x-btuart";
+ clock-indices = <CKEN_FFUART CKEN_BTUART>
+};
+
+
+* Core clock bindings for Marvell PXA27x chips
+
+Required properties:
+- compatible: Should be "marvell,pxa270-core-clocks"
+- #clock-cells: Should be <1>
+- clocks: the parent fixed rate clock, should be the 13MHz oscillator
+- clock-outpout-names: the names of the 7 core clocks
--
2.0.0.rc2
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [PATCH 4/4] clk: dts: document pxa27x clock binding
@ 2014-06-29 18:32 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-29 18:32 UTC (permalink / raw)
To: linux-arm-kernel
Document the device-tree binding of Marvell PXA based SoCs.
PXA clocks are mostly fixed rate and fixed ratio clocks derived from an
external oscillator, and gated by a register set (CKEN or CKEN*).
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
---
.../devicetree/bindings/clock/pxa-clock.txt | 32 ++++++++++++++++++++++
1 file changed, 32 insertions(+)
create mode 100644 Documentation/devicetree/bindings/clock/pxa-clock.txt
diff --git a/Documentation/devicetree/bindings/clock/pxa-clock.txt b/Documentation/devicetree/bindings/clock/pxa-clock.txt
new file mode 100644
index 0000000..beb28ea
--- /dev/null
+++ b/Documentation/devicetree/bindings/clock/pxa-clock.txt
@@ -0,0 +1,32 @@
+* Clock bindings for Marvell PXA chips
+
+Required properties:
+- compatible: Should be "marvell,pxa-clocks"
+- reg: Address and length of the CKEN register
+- #clock-cells: Should be <1>
+- clocks: the parent fixed rate clocks
+- clock-indices: the bit index in the CKEN register gating the clock
+- clock-outpout-names: the names of the clocks
+
+The clock consumer should specify the desired clock by having the clock
+ID in its "clocks" phandle cell.
+
+Examples:
+
+pxa2xx_clks: pxa2xx_clks at 41300004 {
+ compatible = "marvell,pxa-clocks";
+ clocks = <&clk_14_857mhz>, <&clk_14_857mhz>
+ #clock-cells = <1>;
+ clock-output-names =
+ "pxa27x-ffuart", "pxa27x-btuart";
+ clock-indices = <CKEN_FFUART CKEN_BTUART>
+};
+
+
+* Core clock bindings for Marvell PXA27x chips
+
+Required properties:
+- compatible: Should be "marvell,pxa270-core-clocks"
+- #clock-cells: Should be <1>
+- clocks: the parent fixed rate clock, should be the 13MHz oscillator
+- clock-outpout-names: the names of the 7 core clocks
--
2.0.0.rc2
^ permalink raw reply related [flat|nested] 32+ messages in thread
* Re: [PATCH 0/4] Migrate PXA27x platforms to clock framework
2014-06-29 18:32 ` Robert Jarzmik
@ 2014-06-30 6:55 ` Arnd Bergmann
-1 siblings, 0 replies; 32+ messages in thread
From: Arnd Bergmann @ 2014-06-30 6:55 UTC (permalink / raw)
To: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r
Cc: Robert Jarzmik, devicetree-u79uwXL29TY76Z2rM5mHXA,
Mike Turquette, Haojian Zhuang, Eric Miao, Mark Rutland
On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
> As the RFC posted in [1] didn't meet an unrivaled success for
> review, I'm posting this serie for PXA27x transition to clock
> framework.
>
> This transition is needed :
> - to enable device-tree drivers port, as clocks are needed almost
> everywhere
> - to enable the long term multi-platform kernel to support PXA
>
> As I had said before, this serie aims at :
> - keeping legacy platforms working (ie. without device-tree)
> - enable PXA27x to work with a device-tree kernel, and hence
> open the way to drivers conversion
> - be robust enough to support pxa25x and pxa3xx later inclusion
> with almost no change to clk-pxa-dt.c.
>
> As this serie is holding the rest of the device-tree drivers
> port, I'd like it to be reviewed, even it's an old unsexy
> platform.
I have one basic question about this series: if pxa27x gets moved
to used the common-clk framework but the others (pxa25x, pxa26x,
pxa3xx, pxa93x) don't, does that imply that they become mutually
exclusiv at compile-time?
If so, do you plan to first complete all of them before merging
upstream, or do you intend to have one or more kernel releases
that don't allow building a combined kernel for all pxa platforms?
I don't object to doing the latter, but if that is the plan, you
need to make that very clear in the changelog and have all the
relevant maintainers agree to that.
Also (for my understanding) when you say that you plan to do
pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
I assume it does as they are apparently minor revisions of the
former, but it's not completely clear from your description.
Arnd
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-06-30 6:55 ` Arnd Bergmann
0 siblings, 0 replies; 32+ messages in thread
From: Arnd Bergmann @ 2014-06-30 6:55 UTC (permalink / raw)
To: linux-arm-kernel
On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
> As the RFC posted in [1] didn't meet an unrivaled success for
> review, I'm posting this serie for PXA27x transition to clock
> framework.
>
> This transition is needed :
> - to enable device-tree drivers port, as clocks are needed almost
> everywhere
> - to enable the long term multi-platform kernel to support PXA
>
> As I had said before, this serie aims at :
> - keeping legacy platforms working (ie. without device-tree)
> - enable PXA27x to work with a device-tree kernel, and hence
> open the way to drivers conversion
> - be robust enough to support pxa25x and pxa3xx later inclusion
> with almost no change to clk-pxa-dt.c.
>
> As this serie is holding the rest of the device-tree drivers
> port, I'd like it to be reviewed, even it's an old unsexy
> platform.
I have one basic question about this series: if pxa27x gets moved
to used the common-clk framework but the others (pxa25x, pxa26x,
pxa3xx, pxa93x) don't, does that imply that they become mutually
exclusiv at compile-time?
If so, do you plan to first complete all of them before merging
upstream, or do you intend to have one or more kernel releases
that don't allow building a combined kernel for all pxa platforms?
I don't object to doing the latter, but if that is the plan, you
need to make that very clear in the changelog and have all the
relevant maintainers agree to that.
Also (for my understanding) when you say that you plan to do
pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
I assume it does as they are apparently minor revisions of the
former, but it's not completely clear from your description.
Arnd
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 0/4] Migrate PXA27x platforms to clock framework
2014-06-30 6:55 ` Arnd Bergmann
@ 2014-06-30 18:38 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-30 18:38 UTC (permalink / raw)
To: Arnd Bergmann
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Haojian Zhuang, Eric Miao, Mark Rutland
Arnd Bergmann <arnd-r2nGTMty4D4@public.gmane.org> writes:
> On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
>> As the RFC posted in [1] didn't meet an unrivaled success for
>> review, I'm posting this serie for PXA27x transition to clock
>> framework.
>>
>> This transition is needed :
>> - to enable device-tree drivers port, as clocks are needed almost
>> everywhere
>> - to enable the long term multi-platform kernel to support PXA
>>
>> As I had said before, this serie aims at :
>> - keeping legacy platforms working (ie. without device-tree)
>> - enable PXA27x to work with a device-tree kernel, and hence
>> open the way to drivers conversion
>> - be robust enough to support pxa25x and pxa3xx later inclusion
>> with almost no change to clk-pxa-dt.c.
>>
>> As this serie is holding the rest of the device-tree drivers
>> port, I'd like it to be reviewed, even it's an old unsexy
>> platform.
>
> I have one basic question about this series: if pxa27x gets moved
> to used the common-clk framework but the others (pxa25x, pxa26x,
> pxa3xx, pxa93x) don't, does that imply that they become mutually
> exclusiv at compile-time?
Unfortunately yes, they become exclusive.
The reason being that arch/arm/mach-pxa/clock.c defines the function
"clk_enable()", which of course is also defined by the clock framework.
> If so, do you plan to first complete all of them before merging
> upstream, or do you intend to have one or more kernel releases
> that don't allow building a combined kernel for all pxa platforms?
I intend to have first only pxa27x.
Then in a second stage pxa27x + pxa25x + pxa3xx.
> I don't object to doing the latter, but if that is the plan, you
> need to make that very clear in the changelog and have all the
> relevant maintainers agree to that.
OK, that would be Haojian then, I think he maintains all PXA platforms.
Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
platforms even exists (mixing pxa3xx and pxa2xx for example) ?
> Also (for my understanding) when you say that you plan to do
> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
I don't have the Technical Reference Manuals for these ones so the answer is
no. And Google wasn't a great friend at providing them.
> I assume it does as they are apparently minor revisions of the
> former, but it's not completely clear from your description.
My description doesn't mention them, as I have no information about them, nor
any hardware to test on.
Cheers.
--
Robert
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-06-30 18:38 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-06-30 18:38 UTC (permalink / raw)
To: linux-arm-kernel
Arnd Bergmann <arnd@arndb.de> writes:
> On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
>> As the RFC posted in [1] didn't meet an unrivaled success for
>> review, I'm posting this serie for PXA27x transition to clock
>> framework.
>>
>> This transition is needed :
>> - to enable device-tree drivers port, as clocks are needed almost
>> everywhere
>> - to enable the long term multi-platform kernel to support PXA
>>
>> As I had said before, this serie aims at :
>> - keeping legacy platforms working (ie. without device-tree)
>> - enable PXA27x to work with a device-tree kernel, and hence
>> open the way to drivers conversion
>> - be robust enough to support pxa25x and pxa3xx later inclusion
>> with almost no change to clk-pxa-dt.c.
>>
>> As this serie is holding the rest of the device-tree drivers
>> port, I'd like it to be reviewed, even it's an old unsexy
>> platform.
>
> I have one basic question about this series: if pxa27x gets moved
> to used the common-clk framework but the others (pxa25x, pxa26x,
> pxa3xx, pxa93x) don't, does that imply that they become mutually
> exclusiv at compile-time?
Unfortunately yes, they become exclusive.
The reason being that arch/arm/mach-pxa/clock.c defines the function
"clk_enable()", which of course is also defined by the clock framework.
> If so, do you plan to first complete all of them before merging
> upstream, or do you intend to have one or more kernel releases
> that don't allow building a combined kernel for all pxa platforms?
I intend to have first only pxa27x.
Then in a second stage pxa27x + pxa25x + pxa3xx.
> I don't object to doing the latter, but if that is the plan, you
> need to make that very clear in the changelog and have all the
> relevant maintainers agree to that.
OK, that would be Haojian then, I think he maintains all PXA platforms.
Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
platforms even exists (mixing pxa3xx and pxa2xx for example) ?
> Also (for my understanding) when you say that you plan to do
> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
I don't have the Technical Reference Manuals for these ones so the answer is
no. And Google wasn't a great friend at providing them.
> I assume it does as they are apparently minor revisions of the
> former, but it's not completely clear from your description.
My description doesn't mention them, as I have no information about them, nor
any hardware to test on.
Cheers.
--
Robert
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 0/4] Migrate PXA27x platforms to clock framework
2014-06-30 18:38 ` Robert Jarzmik
@ 2014-06-30 20:14 ` Arnd Bergmann
-1 siblings, 0 replies; 32+ messages in thread
From: Arnd Bergmann @ 2014-06-30 20:14 UTC (permalink / raw)
To: Robert Jarzmik
Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Haojian Zhuang, Eric Miao, Mark Rutland
On Monday 30 June 2014 20:38:49 Robert Jarzmik wrote:
> Arnd Bergmann <arnd-r2nGTMty4D4@public.gmane.org> writes:
> > On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
> >> As this serie is holding the rest of the device-tree drivers
> >> port, I'd like it to be reviewed, even it's an old unsexy
> >> platform.
> >
> > I have one basic question about this series: if pxa27x gets moved
> > to used the common-clk framework but the others (pxa25x, pxa26x,
> > pxa3xx, pxa93x) don't, does that imply that they become mutually
> > exclusiv at compile-time?
>
> Unfortunately yes, they become exclusive.
> The reason being that arch/arm/mach-pxa/clock.c defines the function
> "clk_enable()", which of course is also defined by the clock framework.
Ok. Are you able to make it a compile-time decision whether the old
or the new clock code is used?
That way, you can still build everything together (using the old
code) while you are working on the other SoCs.
> > If so, do you plan to first complete all of them before merging
> > upstream, or do you intend to have one or more kernel releases
> > that don't allow building a combined kernel for all pxa platforms?
> I intend to have first only pxa27x.
> Then in a second stage pxa27x + pxa25x + pxa3xx.
>
> > I don't object to doing the latter, but if that is the plan, you
> > need to make that very clear in the changelog and have all the
> > relevant maintainers agree to that.
> OK, that would be Haojian then, I think he maintains all PXA platforms.
>
> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
You can select all machines at compile-time and get a clean build.
I would also assume that it works fine.
> > Also (for my understanding) when you say that you plan to do
> > pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
> I don't have the Technical Reference Manuals for these ones so the answer is
> no. And Google wasn't a great friend at providing them.
I only looked at the current source code, but pxa26x seems to be
basically identical to pxa25x (no relevant code changes at all)
and pxa93x uses the same code as pxa3xx, apparently the only difference
is the addition of the communication processor.
Arnd
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-06-30 20:14 ` Arnd Bergmann
0 siblings, 0 replies; 32+ messages in thread
From: Arnd Bergmann @ 2014-06-30 20:14 UTC (permalink / raw)
To: linux-arm-kernel
On Monday 30 June 2014 20:38:49 Robert Jarzmik wrote:
> Arnd Bergmann <arnd@arndb.de> writes:
> > On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
> >> As this serie is holding the rest of the device-tree drivers
> >> port, I'd like it to be reviewed, even it's an old unsexy
> >> platform.
> >
> > I have one basic question about this series: if pxa27x gets moved
> > to used the common-clk framework but the others (pxa25x, pxa26x,
> > pxa3xx, pxa93x) don't, does that imply that they become mutually
> > exclusiv at compile-time?
>
> Unfortunately yes, they become exclusive.
> The reason being that arch/arm/mach-pxa/clock.c defines the function
> "clk_enable()", which of course is also defined by the clock framework.
Ok. Are you able to make it a compile-time decision whether the old
or the new clock code is used?
That way, you can still build everything together (using the old
code) while you are working on the other SoCs.
> > If so, do you plan to first complete all of them before merging
> > upstream, or do you intend to have one or more kernel releases
> > that don't allow building a combined kernel for all pxa platforms?
> I intend to have first only pxa27x.
> Then in a second stage pxa27x + pxa25x + pxa3xx.
>
> > I don't object to doing the latter, but if that is the plan, you
> > need to make that very clear in the changelog and have all the
> > relevant maintainers agree to that.
> OK, that would be Haojian then, I think he maintains all PXA platforms.
>
> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
You can select all machines at compile-time and get a clean build.
I would also assume that it works fine.
> > Also (for my understanding) when you say that you plan to do
> > pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
> I don't have the Technical Reference Manuals for these ones so the answer is
> no. And Google wasn't a great friend at providing them.
I only looked at the current source code, but pxa26x seems to be
basically identical to pxa25x (no relevant code changes at all)
and pxa93x uses the same code as pxa3xx, apparently the only difference
is the addition of the communication processor.
Arnd
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 1/4] clk: add pxa27x clock drivers
2014-06-29 18:32 ` Robert Jarzmik
@ 2014-07-03 6:12 ` Haojian Zhuang
-1 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-03 6:12 UTC (permalink / raw)
To: Robert Jarzmik
Cc: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette, Eric Miao,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland
On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org> wrote:
> Move pxa27x clock drivers from arch/arm/mach-pxa to driver/clk.
> In the move :
> - convert to new clock framework legacy clocks
> - provide clocks as before for platform data based boards
> - provide clocks through devicetree with clk-pxa-dt
>
> This is the preliminary step in the conversion. The remaining steps are
> :
> - migrate pxa25x and pxa3xx
> - once PXA is fully converted to device tree, if that happens,
> clk-pxa2* and clk-pxa3* should only hold the core clocks which cannot
> be described in devicetree.
>
> Signed-off-by: Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>
> ---
> drivers/clk/Makefile | 1 +
> drivers/clk/pxa/Makefile | 4 +
> drivers/clk/pxa/clk-pxa-dt.c | 76 ++++++++++
> drivers/clk/pxa/clk-pxa27x.c | 324 +++++++++++++++++++++++++++++++++++++++++++
> drivers/clk/pxa/clk-pxa2xx.c | 74 ++++++++++
> drivers/clk/pxa/clk-pxa2xx.h | 47 +++++++
> 6 files changed, 526 insertions(+)
> create mode 100644 drivers/clk/pxa/Makefile
> create mode 100644 drivers/clk/pxa/clk-pxa-dt.c
> create mode 100644 drivers/clk/pxa/clk-pxa27x.c
> create mode 100644 drivers/clk/pxa/clk-pxa2xx.c
> create mode 100644 drivers/clk/pxa/clk-pxa2xx.h
>
> +
> +static struct pxa27x_clocks_fixed_cken pxa27x_without_dt_clocks[] __initdata = {
> + PXA2_FIXED_RATE("pxa2xx-uart.0", NULL, FFUART, 14857000, 1),
> + PXA2_FIXED_RATE("pxa2xx-uart.1", NULL, BTUART, 14857000, 1),
> + PXA2_FIXED_RATE("pxa2xx-uart.2", NULL, STUART, 14857000, 1),
> + PXA2_FIXED_RATE(NULL, "UARTCLK", STUART, 14857000, 1),
> + PXA2_FIXED_RATE("pxa2xx-i2s", NULL, I2S, 14682000, 0),
> + PXA2_FIXED_RATE("pxa2xx-i2c.0", NULL, I2C, 32842000, 0),
> + PXA2_FIXED_RATE("pxa27x-udc", NULL, USB, 48000000, 5),
> + PXA2_FIXED_RATE("pxa2xx-mci.0", NULL, MMC, 19500000, 0),
> + PXA2_FIXED_RATE("pxa2xx-ir", "FICPCLK", FICP, 48000000, 0),
> + PXA2_FIXED_RATE("pxa27x-ohci", NULL, USBHOST, 48000000, 0),
> + PXA2_FIXED_RATE("pxa2xx-i2c.1", NULL, PWRI2C, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-keypad", NULL, KEYPAD, 32768, 0),
> + PXA2_FIXED_RATE("pxa27x-ssp.0", NULL, SSP1, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-ssp.1", NULL, SSP2, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-ssp.2", NULL, SSP3, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-pwm.0", NULL, PWM0, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-pwm.1", NULL, PWM1, 13000000, 0),
> + PXA2_FIXED_RATE(NULL, "AC97CLK", AC97, 24576000, 0),
> + PXA2_FIXED_RATE(NULL, "AC97CONFCLK", AC97CONF, 24576000, 0),
> + PXA2_FIXED_RATE(NULL, "MSLCLK", MSL, 48000000, 0),
> + PXA2_FIXED_RATE(NULL, "USIMCLK", USIM, 48000000, 0),
> + PXA2_FIXED_RATE(NULL, "MSTKCLK", MEMSTK, 19500000, 0),
> + PXA2_FIXED_RATE(NULL, "IMCLK", IM, 0, 0),
> + PXA2_FIXED_RATE_AO("pxa27x-memc", "MEMCLK", MEMC, 0, 0),
> +};
> +
> +static struct pxa27x_clocks_var_rate pxa27x_var_rate_clocks[] __initdata = {
> + PXA2_VAR_RATE("pxa2xx-fb", true, LCD, &clk_pxa27x_lcd_ops),
> + PXA2_VAR_RATE("pxa27x-camera.0", true, CAMERA, &clk_pxa27x_lcd_ops),
> + PXA2_VAR_RATE("pxa2xx-pcmcia", false, MEMC, &clk_pxa27x_mem_ops),
> +};
> +
It's not good to define a VAR table any more.
Now we have the common clock framework. Since these three clocks are
used as clock
gate. And their clock frequency is controlled by other bits in
peripheral controller registers.
We can define parent & child clocks in clock tables. I think that you
can get a lot of
reference in clock drivers, such as exynos, hisilicon, ...
Same comment on PXA_FIXED_RATE table.
Regards
Haojian
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 1/4] clk: add pxa27x clock drivers
@ 2014-07-03 6:12 ` Haojian Zhuang
0 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-03 6:12 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik@free.fr> wrote:
> Move pxa27x clock drivers from arch/arm/mach-pxa to driver/clk.
> In the move :
> - convert to new clock framework legacy clocks
> - provide clocks as before for platform data based boards
> - provide clocks through devicetree with clk-pxa-dt
>
> This is the preliminary step in the conversion. The remaining steps are
> :
> - migrate pxa25x and pxa3xx
> - once PXA is fully converted to device tree, if that happens,
> clk-pxa2* and clk-pxa3* should only hold the core clocks which cannot
> be described in devicetree.
>
> Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
> ---
> drivers/clk/Makefile | 1 +
> drivers/clk/pxa/Makefile | 4 +
> drivers/clk/pxa/clk-pxa-dt.c | 76 ++++++++++
> drivers/clk/pxa/clk-pxa27x.c | 324 +++++++++++++++++++++++++++++++++++++++++++
> drivers/clk/pxa/clk-pxa2xx.c | 74 ++++++++++
> drivers/clk/pxa/clk-pxa2xx.h | 47 +++++++
> 6 files changed, 526 insertions(+)
> create mode 100644 drivers/clk/pxa/Makefile
> create mode 100644 drivers/clk/pxa/clk-pxa-dt.c
> create mode 100644 drivers/clk/pxa/clk-pxa27x.c
> create mode 100644 drivers/clk/pxa/clk-pxa2xx.c
> create mode 100644 drivers/clk/pxa/clk-pxa2xx.h
>
> +
> +static struct pxa27x_clocks_fixed_cken pxa27x_without_dt_clocks[] __initdata = {
> + PXA2_FIXED_RATE("pxa2xx-uart.0", NULL, FFUART, 14857000, 1),
> + PXA2_FIXED_RATE("pxa2xx-uart.1", NULL, BTUART, 14857000, 1),
> + PXA2_FIXED_RATE("pxa2xx-uart.2", NULL, STUART, 14857000, 1),
> + PXA2_FIXED_RATE(NULL, "UARTCLK", STUART, 14857000, 1),
> + PXA2_FIXED_RATE("pxa2xx-i2s", NULL, I2S, 14682000, 0),
> + PXA2_FIXED_RATE("pxa2xx-i2c.0", NULL, I2C, 32842000, 0),
> + PXA2_FIXED_RATE("pxa27x-udc", NULL, USB, 48000000, 5),
> + PXA2_FIXED_RATE("pxa2xx-mci.0", NULL, MMC, 19500000, 0),
> + PXA2_FIXED_RATE("pxa2xx-ir", "FICPCLK", FICP, 48000000, 0),
> + PXA2_FIXED_RATE("pxa27x-ohci", NULL, USBHOST, 48000000, 0),
> + PXA2_FIXED_RATE("pxa2xx-i2c.1", NULL, PWRI2C, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-keypad", NULL, KEYPAD, 32768, 0),
> + PXA2_FIXED_RATE("pxa27x-ssp.0", NULL, SSP1, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-ssp.1", NULL, SSP2, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-ssp.2", NULL, SSP3, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-pwm.0", NULL, PWM0, 13000000, 0),
> + PXA2_FIXED_RATE("pxa27x-pwm.1", NULL, PWM1, 13000000, 0),
> + PXA2_FIXED_RATE(NULL, "AC97CLK", AC97, 24576000, 0),
> + PXA2_FIXED_RATE(NULL, "AC97CONFCLK", AC97CONF, 24576000, 0),
> + PXA2_FIXED_RATE(NULL, "MSLCLK", MSL, 48000000, 0),
> + PXA2_FIXED_RATE(NULL, "USIMCLK", USIM, 48000000, 0),
> + PXA2_FIXED_RATE(NULL, "MSTKCLK", MEMSTK, 19500000, 0),
> + PXA2_FIXED_RATE(NULL, "IMCLK", IM, 0, 0),
> + PXA2_FIXED_RATE_AO("pxa27x-memc", "MEMCLK", MEMC, 0, 0),
> +};
> +
> +static struct pxa27x_clocks_var_rate pxa27x_var_rate_clocks[] __initdata = {
> + PXA2_VAR_RATE("pxa2xx-fb", true, LCD, &clk_pxa27x_lcd_ops),
> + PXA2_VAR_RATE("pxa27x-camera.0", true, CAMERA, &clk_pxa27x_lcd_ops),
> + PXA2_VAR_RATE("pxa2xx-pcmcia", false, MEMC, &clk_pxa27x_mem_ops),
> +};
> +
It's not good to define a VAR table any more.
Now we have the common clock framework. Since these three clocks are
used as clock
gate. And their clock frequency is controlled by other bits in
peripheral controller registers.
We can define parent & child clocks in clock tables. I think that you
can get a lot of
reference in clock drivers, such as exynos, hisilicon, ...
Same comment on PXA_FIXED_RATE table.
Regards
Haojian
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
2014-06-29 18:32 ` Robert Jarzmik
@ 2014-07-03 6:14 ` Haojian Zhuang
-1 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-03 6:14 UTC (permalink / raw)
To: Robert Jarzmik
Cc: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette, Eric Miao,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland
On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org> wrote:
> Add the clock tree description for the PXA27x based boards.
>
> Signed-off-by: Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>
> ---
> arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++++++++++++++++++++++++-
> include/dt-bindings/clock/pxa2xx-clock.h | 45 +++++++++++
> 2 files changed, 178 insertions(+), 1 deletion(-)
> create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
>
> diff --git a/arch/arm/boot/dts/pxa27x.dtsi b/arch/arm/boot/dts/pxa27x.dtsi
> index a705469..badaa71 100644
> --- a/arch/arm/boot/dts/pxa27x.dtsi
> +++ b/arch/arm/boot/dts/pxa27x.dtsi
> @@ -1,5 +1,6 @@
> /* The pxa3xx skeleton simply augments the 2xx version */
> -/include/ "pxa2xx.dtsi"
> +#include "pxa2xx.dtsi"
> +#include "dt-bindings/clock/pxa2xx-clock.h"
>
> / {
> model = "Marvell PXA27x familiy SoC";
> @@ -35,4 +36,135 @@
> #pwm-cells = <1>;
> };
> };
> +
> + clocks {
> + /*
> + * The muxing of external clocks/internal dividers for osc* clock
> + * sources has been hidden under the carpet by now.
> + */
> + #address-cells = <1>;
> + #size-cells = <1>;
> + ranges;
> +
> + osc13mhz:osc13mhz {
> + compatible = "fixed-clock";
> + #clock-cells = <0>;
> + clock-frequency = <13000000>;
> + };
> +
> + osc32_768khz:osc32_768khz {
> + compatible = "fixed-clock";
> + #clock-cells = <0>;
> + clock-frequency = <32768>;
> + };
> +
> + pll_312mhz:pll_312mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&osc13mhz>;
> + clock-div = <1>;
> + clock-mult = <24>;
> + };
> +
> + clk_48mhz:clk_48mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <13>;
> + clock-mult = <2>;
> + };
> +
> + clk_32_842mhz:clk_32_842mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <19>;
> + clock-mult = <2>;
> + };
> +
> + clk_19_5mhz:clk_19_5mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <32>;
> + clock-mult = <2>;
> + };
> +
> + clk_14_857mhz:clk_14_857mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <42>;
> + clock-mult = <2>;
> + };
> +
> + clk_14_682mhz:clk_14_682mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <51>;
> + clock-mult = <2>;
> + };
> +
> + clk_13mhz:clk_13mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&osc13mhz>;
> + clock-div = <1>;
> + clock-mult = <1>;
> + };
> +
> + clk_dummy:clk_dummy {
> + compatible = "fixed-clock";
> + #clock-cells = <0>;
> + clock-frequency = <0>;
> + };
> +
> + clk_ostimer:clk_ostimer {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&osc13mhz>;
> + clock-div = <4>;
> + clock-mult = <1>;
> + };
> +
> + pxa27x_sysclks:pxa27x_sysclks {
> + compatible = "marvell,pxa270-core-clocks";
> + #clock-cells = <1>;
> + clocks = <&osc13mhz>;
> + clock-output-names = "run mode", "half-turbo mode",
> + "turbo mode", "cpu core", "system bus", "memory", "lcd";
> + };
> +
> + pxa2xx_clks: pxa2xx_clks@41300004 {
> + compatible = "marvell,pxa-clocks";
> + reg = <0x41300004 0x4>;
> + clocks =
> + <&clk_13mhz>, <&clk_13mhz>, <&clk_dummy>, <&clk_13mhz>,
> + <&clk_13mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>,
> + <&clk_14_682mhz>, <&clk_ostimer>, <&clk_48mhz>, <&clk_48mhz>,
> + <&clk_19_5mhz>, <&clk_48mhz>, <&clk_32_842mhz>, <&clk_13mhz>,
> + <&pxa27x_sysclks 6>, <&clk_48mhz>, <&clk_48mhz>, <&osc32_768khz>,
> + <&clk_dummy>, <&clk_19_5mhz>, <&pxa27x_sysclks 4>, <&clk_13mhz>,
> + <&pxa27x_sysclks 6>, <&clk_dummy>;
> + #clock-cells = <1>;
> + clock-output-names =
> + "pwm 0,2", "pwm 1,3", "ac97", "ssp2",
> + "ssp3,hwuart", "stuart", "ffuart", "btuart",
> + "i2s", "nssp,ostimer", "usb host,assp", "usb udc",
> + "mmc", "ficp", "i2c", "pwri2c",
> + "lcd", "msl", "usim", "keypad",
> + "im", "memstk", "memc", "ssp1",
> + "camera", "ac97conf";
> + clock-indices = <
> + CKEN_PWM0 CKEN_PWM1 CKEN_AC97 CKEN_SSP2
> + CKEN_SSP3 CKEN_STUART CKEN_FFUART CKEN_BTUART
> + CKEN_I2S CKEN_OSTIMER CKEN_USBHOST CKEN_USB
> + CKEN_MMC CKEN_FICP CKEN_I2C CKEN_PWRI2C
> + CKEN_LCD CKEN_MSL CKEN_USIM CKEN_KEYPAD
> + CKEN_IM CKEN_MEMSTK 65 CKEN_SSP1
> + CKEN_CAMERA CKEN_AC97CONF >;
> + };
> + };
> +
> };
Maybe defining these clocks in pxa27x clock driver is better.
Regards
Haojian
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
@ 2014-07-03 6:14 ` Haojian Zhuang
0 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-03 6:14 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik@free.fr> wrote:
> Add the clock tree description for the PXA27x based boards.
>
> Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
> ---
> arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++++++++++++++++++++++++-
> include/dt-bindings/clock/pxa2xx-clock.h | 45 +++++++++++
> 2 files changed, 178 insertions(+), 1 deletion(-)
> create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
>
> diff --git a/arch/arm/boot/dts/pxa27x.dtsi b/arch/arm/boot/dts/pxa27x.dtsi
> index a705469..badaa71 100644
> --- a/arch/arm/boot/dts/pxa27x.dtsi
> +++ b/arch/arm/boot/dts/pxa27x.dtsi
> @@ -1,5 +1,6 @@
> /* The pxa3xx skeleton simply augments the 2xx version */
> -/include/ "pxa2xx.dtsi"
> +#include "pxa2xx.dtsi"
> +#include "dt-bindings/clock/pxa2xx-clock.h"
>
> / {
> model = "Marvell PXA27x familiy SoC";
> @@ -35,4 +36,135 @@
> #pwm-cells = <1>;
> };
> };
> +
> + clocks {
> + /*
> + * The muxing of external clocks/internal dividers for osc* clock
> + * sources has been hidden under the carpet by now.
> + */
> + #address-cells = <1>;
> + #size-cells = <1>;
> + ranges;
> +
> + osc13mhz:osc13mhz {
> + compatible = "fixed-clock";
> + #clock-cells = <0>;
> + clock-frequency = <13000000>;
> + };
> +
> + osc32_768khz:osc32_768khz {
> + compatible = "fixed-clock";
> + #clock-cells = <0>;
> + clock-frequency = <32768>;
> + };
> +
> + pll_312mhz:pll_312mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&osc13mhz>;
> + clock-div = <1>;
> + clock-mult = <24>;
> + };
> +
> + clk_48mhz:clk_48mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <13>;
> + clock-mult = <2>;
> + };
> +
> + clk_32_842mhz:clk_32_842mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <19>;
> + clock-mult = <2>;
> + };
> +
> + clk_19_5mhz:clk_19_5mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <32>;
> + clock-mult = <2>;
> + };
> +
> + clk_14_857mhz:clk_14_857mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <42>;
> + clock-mult = <2>;
> + };
> +
> + clk_14_682mhz:clk_14_682mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&pll_312mhz>;
> + clock-div = <51>;
> + clock-mult = <2>;
> + };
> +
> + clk_13mhz:clk_13mhz {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&osc13mhz>;
> + clock-div = <1>;
> + clock-mult = <1>;
> + };
> +
> + clk_dummy:clk_dummy {
> + compatible = "fixed-clock";
> + #clock-cells = <0>;
> + clock-frequency = <0>;
> + };
> +
> + clk_ostimer:clk_ostimer {
> + compatible = "fixed-factor-clock";
> + #clock-cells = <0>;
> + clocks = <&osc13mhz>;
> + clock-div = <4>;
> + clock-mult = <1>;
> + };
> +
> + pxa27x_sysclks:pxa27x_sysclks {
> + compatible = "marvell,pxa270-core-clocks";
> + #clock-cells = <1>;
> + clocks = <&osc13mhz>;
> + clock-output-names = "run mode", "half-turbo mode",
> + "turbo mode", "cpu core", "system bus", "memory", "lcd";
> + };
> +
> + pxa2xx_clks: pxa2xx_clks at 41300004 {
> + compatible = "marvell,pxa-clocks";
> + reg = <0x41300004 0x4>;
> + clocks =
> + <&clk_13mhz>, <&clk_13mhz>, <&clk_dummy>, <&clk_13mhz>,
> + <&clk_13mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>,
> + <&clk_14_682mhz>, <&clk_ostimer>, <&clk_48mhz>, <&clk_48mhz>,
> + <&clk_19_5mhz>, <&clk_48mhz>, <&clk_32_842mhz>, <&clk_13mhz>,
> + <&pxa27x_sysclks 6>, <&clk_48mhz>, <&clk_48mhz>, <&osc32_768khz>,
> + <&clk_dummy>, <&clk_19_5mhz>, <&pxa27x_sysclks 4>, <&clk_13mhz>,
> + <&pxa27x_sysclks 6>, <&clk_dummy>;
> + #clock-cells = <1>;
> + clock-output-names =
> + "pwm 0,2", "pwm 1,3", "ac97", "ssp2",
> + "ssp3,hwuart", "stuart", "ffuart", "btuart",
> + "i2s", "nssp,ostimer", "usb host,assp", "usb udc",
> + "mmc", "ficp", "i2c", "pwri2c",
> + "lcd", "msl", "usim", "keypad",
> + "im", "memstk", "memc", "ssp1",
> + "camera", "ac97conf";
> + clock-indices = <
> + CKEN_PWM0 CKEN_PWM1 CKEN_AC97 CKEN_SSP2
> + CKEN_SSP3 CKEN_STUART CKEN_FFUART CKEN_BTUART
> + CKEN_I2S CKEN_OSTIMER CKEN_USBHOST CKEN_USB
> + CKEN_MMC CKEN_FICP CKEN_I2C CKEN_PWRI2C
> + CKEN_LCD CKEN_MSL CKEN_USIM CKEN_KEYPAD
> + CKEN_IM CKEN_MEMSTK 65 CKEN_SSP1
> + CKEN_CAMERA CKEN_AC97CONF >;
> + };
> + };
> +
> };
Maybe defining these clocks in pxa27x clock driver is better.
Regards
Haojian
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 0/4] Migrate PXA27x platforms to clock framework
2014-06-30 18:38 ` Robert Jarzmik
@ 2014-07-03 6:21 ` Haojian Zhuang
-1 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-03 6:21 UTC (permalink / raw)
To: Robert Jarzmik
Cc: Arnd Bergmann, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette, Eric Miao,
Mark Rutland
On Tue, Jul 1, 2014 at 2:38 AM, Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org> wrote:
> Arnd Bergmann <arnd-r2nGTMty4D4@public.gmane.org> writes:
>
>> On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
>>> As the RFC posted in [1] didn't meet an unrivaled success for
>>> review, I'm posting this serie for PXA27x transition to clock
>>> framework.
>>>
>>> This transition is needed :
>>> - to enable device-tree drivers port, as clocks are needed almost
>>> everywhere
>>> - to enable the long term multi-platform kernel to support PXA
>>>
>>> As I had said before, this serie aims at :
>>> - keeping legacy platforms working (ie. without device-tree)
>>> - enable PXA27x to work with a device-tree kernel, and hence
>>> open the way to drivers conversion
>>> - be robust enough to support pxa25x and pxa3xx later inclusion
>>> with almost no change to clk-pxa-dt.c.
>>>
>>> As this serie is holding the rest of the device-tree drivers
>>> port, I'd like it to be reviewed, even it's an old unsexy
>>> platform.
>>
>> I have one basic question about this series: if pxa27x gets moved
>> to used the common-clk framework but the others (pxa25x, pxa26x,
>> pxa3xx, pxa93x) don't, does that imply that they become mutually
>> exclusiv at compile-time?
>
> Unfortunately yes, they become exclusive.
> The reason being that arch/arm/mach-pxa/clock.c defines the function
> "clk_enable()", which of course is also defined by the clock framework.
>
>> If so, do you plan to first complete all of them before merging
>> upstream, or do you intend to have one or more kernel releases
>> that don't allow building a combined kernel for all pxa platforms?
> I intend to have first only pxa27x.
> Then in a second stage pxa27x + pxa25x + pxa3xx.
>
>> I don't object to doing the latter, but if that is the plan, you
>> need to make that very clear in the changelog and have all the
>> relevant maintainers agree to that.
> OK, that would be Haojian then, I think he maintains all PXA platforms.
>
> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
>
It's acceptable to me that different silicons are queued in different stages.
I only request that it won't break the compiler building & bootup.
But I think that the pxa clock driver may be shared among all PXA silicons
except for the clock table. What's your opinion?
>> Also (for my understanding) when you say that you plan to do
>> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
> I don't have the Technical Reference Manuals for these ones so the answer is
> no. And Google wasn't a great friend at providing them.
>
Converting them into new clock driver may not rely on the reference manual.
>> I assume it does as they are apparently minor revisions of the
>> former, but it's not completely clear from your description.
> My description doesn't mention them, as I have no information about them, nor
> any hardware to test on.
>
We can request others to help testing in the mailing list.
Regards
Haojian
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-07-03 6:21 ` Haojian Zhuang
0 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-03 6:21 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, Jul 1, 2014 at 2:38 AM, Robert Jarzmik <robert.jarzmik@free.fr> wrote:
> Arnd Bergmann <arnd@arndb.de> writes:
>
>> On Sunday 29 June 2014 20:32:20 Robert Jarzmik wrote:
>>> As the RFC posted in [1] didn't meet an unrivaled success for
>>> review, I'm posting this serie for PXA27x transition to clock
>>> framework.
>>>
>>> This transition is needed :
>>> - to enable device-tree drivers port, as clocks are needed almost
>>> everywhere
>>> - to enable the long term multi-platform kernel to support PXA
>>>
>>> As I had said before, this serie aims at :
>>> - keeping legacy platforms working (ie. without device-tree)
>>> - enable PXA27x to work with a device-tree kernel, and hence
>>> open the way to drivers conversion
>>> - be robust enough to support pxa25x and pxa3xx later inclusion
>>> with almost no change to clk-pxa-dt.c.
>>>
>>> As this serie is holding the rest of the device-tree drivers
>>> port, I'd like it to be reviewed, even it's an old unsexy
>>> platform.
>>
>> I have one basic question about this series: if pxa27x gets moved
>> to used the common-clk framework but the others (pxa25x, pxa26x,
>> pxa3xx, pxa93x) don't, does that imply that they become mutually
>> exclusiv at compile-time?
>
> Unfortunately yes, they become exclusive.
> The reason being that arch/arm/mach-pxa/clock.c defines the function
> "clk_enable()", which of course is also defined by the clock framework.
>
>> If so, do you plan to first complete all of them before merging
>> upstream, or do you intend to have one or more kernel releases
>> that don't allow building a combined kernel for all pxa platforms?
> I intend to have first only pxa27x.
> Then in a second stage pxa27x + pxa25x + pxa3xx.
>
>> I don't object to doing the latter, but if that is the plan, you
>> need to make that very clear in the changelog and have all the
>> relevant maintainers agree to that.
> OK, that would be Haojian then, I think he maintains all PXA platforms.
>
> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
>
It's acceptable to me that different silicons are queued in different stages.
I only request that it won't break the compiler building & bootup.
But I think that the pxa clock driver may be shared among all PXA silicons
except for the clock table. What's your opinion?
>> Also (for my understanding) when you say that you plan to do
>> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
> I don't have the Technical Reference Manuals for these ones so the answer is
> no. And Google wasn't a great friend at providing them.
>
Converting them into new clock driver may not rely on the reference manual.
>> I assume it does as they are apparently minor revisions of the
>> former, but it's not completely clear from your description.
> My description doesn't mention them, as I have no information about them, nor
> any hardware to test on.
>
We can request others to help testing in the mailing list.
Regards
Haojian
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
2014-07-03 6:14 ` Haojian Zhuang
@ 2014-07-03 22:03 ` Mike Turquette
-1 siblings, 0 replies; 32+ messages in thread
From: Mike Turquette @ 2014-07-03 22:03 UTC (permalink / raw)
To: Haojian Zhuang, Robert Jarzmik
Cc: devicetree-u79uwXL29TY76Z2rM5mHXA, Eric Miao,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Mark Rutland
Quoting Haojian Zhuang (2014-07-02 23:14:33)
> On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org> wrote:
> > Add the clock tree description for the PXA27x based boards.
> >
> > Signed-off-by: Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>
> > ---
> > arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++++++++++++++++++++++++-
> > include/dt-bindings/clock/pxa2xx-clock.h | 45 +++++++++++
> > 2 files changed, 178 insertions(+), 1 deletion(-)
> > create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
> >
> > diff --git a/arch/arm/boot/dts/pxa27x.dtsi b/arch/arm/boot/dts/pxa27x.dtsi
> > index a705469..badaa71 100644
> > --- a/arch/arm/boot/dts/pxa27x.dtsi
> > +++ b/arch/arm/boot/dts/pxa27x.dtsi
> > @@ -1,5 +1,6 @@
> > /* The pxa3xx skeleton simply augments the 2xx version */
> > -/include/ "pxa2xx.dtsi"
> > +#include "pxa2xx.dtsi"
> > +#include "dt-bindings/clock/pxa2xx-clock.h"
> >
> > / {
> > model = "Marvell PXA27x familiy SoC";
> > @@ -35,4 +36,135 @@
> > #pwm-cells = <1>;
> > };
> > };
> > +
> > + clocks {
> > + /*
> > + * The muxing of external clocks/internal dividers for osc* clock
> > + * sources has been hidden under the carpet by now.
> > + */
> > + #address-cells = <1>;
> > + #size-cells = <1>;
> > + ranges;
> > +
> > + osc13mhz:osc13mhz {
> > + compatible = "fixed-clock";
> > + #clock-cells = <0>;
> > + clock-frequency = <13000000>;
> > + };
> > +
> > + osc32_768khz:osc32_768khz {
> > + compatible = "fixed-clock";
> > + #clock-cells = <0>;
> > + clock-frequency = <32768>;
> > + };
> > +
> > + pll_312mhz:pll_312mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&osc13mhz>;
> > + clock-div = <1>;
> > + clock-mult = <24>;
> > + };
> > +
> > + clk_48mhz:clk_48mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <13>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_32_842mhz:clk_32_842mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <19>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_19_5mhz:clk_19_5mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <32>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_14_857mhz:clk_14_857mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <42>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_14_682mhz:clk_14_682mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <51>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_13mhz:clk_13mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&osc13mhz>;
> > + clock-div = <1>;
> > + clock-mult = <1>;
> > + };
> > +
> > + clk_dummy:clk_dummy {
> > + compatible = "fixed-clock";
> > + #clock-cells = <0>;
> > + clock-frequency = <0>;
> > + };
> > +
> > + clk_ostimer:clk_ostimer {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&osc13mhz>;
> > + clock-div = <4>;
> > + clock-mult = <1>;
> > + };
> > +
> > + pxa27x_sysclks:pxa27x_sysclks {
> > + compatible = "marvell,pxa270-core-clocks";
> > + #clock-cells = <1>;
> > + clocks = <&osc13mhz>;
> > + clock-output-names = "run mode", "half-turbo mode",
> > + "turbo mode", "cpu core", "system bus", "memory", "lcd";
> > + };
> > +
> > + pxa2xx_clks: pxa2xx_clks@41300004 {
> > + compatible = "marvell,pxa-clocks";
> > + reg = <0x41300004 0x4>;
> > + clocks =
> > + <&clk_13mhz>, <&clk_13mhz>, <&clk_dummy>, <&clk_13mhz>,
> > + <&clk_13mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>,
> > + <&clk_14_682mhz>, <&clk_ostimer>, <&clk_48mhz>, <&clk_48mhz>,
> > + <&clk_19_5mhz>, <&clk_48mhz>, <&clk_32_842mhz>, <&clk_13mhz>,
> > + <&pxa27x_sysclks 6>, <&clk_48mhz>, <&clk_48mhz>, <&osc32_768khz>,
> > + <&clk_dummy>, <&clk_19_5mhz>, <&pxa27x_sysclks 4>, <&clk_13mhz>,
> > + <&pxa27x_sysclks 6>, <&clk_dummy>;
> > + #clock-cells = <1>;
> > + clock-output-names =
> > + "pwm 0,2", "pwm 1,3", "ac97", "ssp2",
> > + "ssp3,hwuart", "stuart", "ffuart", "btuart",
> > + "i2s", "nssp,ostimer", "usb host,assp", "usb udc",
> > + "mmc", "ficp", "i2c", "pwri2c",
> > + "lcd", "msl", "usim", "keypad",
> > + "im", "memstk", "memc", "ssp1",
> > + "camera", "ac97conf";
> > + clock-indices = <
> > + CKEN_PWM0 CKEN_PWM1 CKEN_AC97 CKEN_SSP2
> > + CKEN_SSP3 CKEN_STUART CKEN_FFUART CKEN_BTUART
> > + CKEN_I2S CKEN_OSTIMER CKEN_USBHOST CKEN_USB
> > + CKEN_MMC CKEN_FICP CKEN_I2C CKEN_PWRI2C
> > + CKEN_LCD CKEN_MSL CKEN_USIM CKEN_KEYPAD
> > + CKEN_IM CKEN_MEMSTK 65 CKEN_SSP1
> > + CKEN_CAMERA CKEN_AC97CONF >;
> > + };
> > + };
> > +
> > };
>
> Maybe defining these clocks in pxa27x clock driver is better.
Agreed. After a long time trying to figure out the best DT binding for
clocks it seems that defining the per-clock data in DTS is not the best
way. The qcom msm and samsung exynos clock bindings and DTS data are
good examples.
Regards,
Mike
>
> Regards
> Haojian
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
@ 2014-07-03 22:03 ` Mike Turquette
0 siblings, 0 replies; 32+ messages in thread
From: Mike Turquette @ 2014-07-03 22:03 UTC (permalink / raw)
To: linux-arm-kernel
Quoting Haojian Zhuang (2014-07-02 23:14:33)
> On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik@free.fr> wrote:
> > Add the clock tree description for the PXA27x based boards.
> >
> > Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
> > ---
> > arch/arm/boot/dts/pxa27x.dtsi | 134 ++++++++++++++++++++++++++++++-
> > include/dt-bindings/clock/pxa2xx-clock.h | 45 +++++++++++
> > 2 files changed, 178 insertions(+), 1 deletion(-)
> > create mode 100644 include/dt-bindings/clock/pxa2xx-clock.h
> >
> > diff --git a/arch/arm/boot/dts/pxa27x.dtsi b/arch/arm/boot/dts/pxa27x.dtsi
> > index a705469..badaa71 100644
> > --- a/arch/arm/boot/dts/pxa27x.dtsi
> > +++ b/arch/arm/boot/dts/pxa27x.dtsi
> > @@ -1,5 +1,6 @@
> > /* The pxa3xx skeleton simply augments the 2xx version */
> > -/include/ "pxa2xx.dtsi"
> > +#include "pxa2xx.dtsi"
> > +#include "dt-bindings/clock/pxa2xx-clock.h"
> >
> > / {
> > model = "Marvell PXA27x familiy SoC";
> > @@ -35,4 +36,135 @@
> > #pwm-cells = <1>;
> > };
> > };
> > +
> > + clocks {
> > + /*
> > + * The muxing of external clocks/internal dividers for osc* clock
> > + * sources has been hidden under the carpet by now.
> > + */
> > + #address-cells = <1>;
> > + #size-cells = <1>;
> > + ranges;
> > +
> > + osc13mhz:osc13mhz {
> > + compatible = "fixed-clock";
> > + #clock-cells = <0>;
> > + clock-frequency = <13000000>;
> > + };
> > +
> > + osc32_768khz:osc32_768khz {
> > + compatible = "fixed-clock";
> > + #clock-cells = <0>;
> > + clock-frequency = <32768>;
> > + };
> > +
> > + pll_312mhz:pll_312mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&osc13mhz>;
> > + clock-div = <1>;
> > + clock-mult = <24>;
> > + };
> > +
> > + clk_48mhz:clk_48mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <13>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_32_842mhz:clk_32_842mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <19>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_19_5mhz:clk_19_5mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <32>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_14_857mhz:clk_14_857mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <42>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_14_682mhz:clk_14_682mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&pll_312mhz>;
> > + clock-div = <51>;
> > + clock-mult = <2>;
> > + };
> > +
> > + clk_13mhz:clk_13mhz {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&osc13mhz>;
> > + clock-div = <1>;
> > + clock-mult = <1>;
> > + };
> > +
> > + clk_dummy:clk_dummy {
> > + compatible = "fixed-clock";
> > + #clock-cells = <0>;
> > + clock-frequency = <0>;
> > + };
> > +
> > + clk_ostimer:clk_ostimer {
> > + compatible = "fixed-factor-clock";
> > + #clock-cells = <0>;
> > + clocks = <&osc13mhz>;
> > + clock-div = <4>;
> > + clock-mult = <1>;
> > + };
> > +
> > + pxa27x_sysclks:pxa27x_sysclks {
> > + compatible = "marvell,pxa270-core-clocks";
> > + #clock-cells = <1>;
> > + clocks = <&osc13mhz>;
> > + clock-output-names = "run mode", "half-turbo mode",
> > + "turbo mode", "cpu core", "system bus", "memory", "lcd";
> > + };
> > +
> > + pxa2xx_clks: pxa2xx_clks at 41300004 {
> > + compatible = "marvell,pxa-clocks";
> > + reg = <0x41300004 0x4>;
> > + clocks =
> > + <&clk_13mhz>, <&clk_13mhz>, <&clk_dummy>, <&clk_13mhz>,
> > + <&clk_13mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>, <&clk_14_857mhz>,
> > + <&clk_14_682mhz>, <&clk_ostimer>, <&clk_48mhz>, <&clk_48mhz>,
> > + <&clk_19_5mhz>, <&clk_48mhz>, <&clk_32_842mhz>, <&clk_13mhz>,
> > + <&pxa27x_sysclks 6>, <&clk_48mhz>, <&clk_48mhz>, <&osc32_768khz>,
> > + <&clk_dummy>, <&clk_19_5mhz>, <&pxa27x_sysclks 4>, <&clk_13mhz>,
> > + <&pxa27x_sysclks 6>, <&clk_dummy>;
> > + #clock-cells = <1>;
> > + clock-output-names =
> > + "pwm 0,2", "pwm 1,3", "ac97", "ssp2",
> > + "ssp3,hwuart", "stuart", "ffuart", "btuart",
> > + "i2s", "nssp,ostimer", "usb host,assp", "usb udc",
> > + "mmc", "ficp", "i2c", "pwri2c",
> > + "lcd", "msl", "usim", "keypad",
> > + "im", "memstk", "memc", "ssp1",
> > + "camera", "ac97conf";
> > + clock-indices = <
> > + CKEN_PWM0 CKEN_PWM1 CKEN_AC97 CKEN_SSP2
> > + CKEN_SSP3 CKEN_STUART CKEN_FFUART CKEN_BTUART
> > + CKEN_I2S CKEN_OSTIMER CKEN_USBHOST CKEN_USB
> > + CKEN_MMC CKEN_FICP CKEN_I2C CKEN_PWRI2C
> > + CKEN_LCD CKEN_MSL CKEN_USIM CKEN_KEYPAD
> > + CKEN_IM CKEN_MEMSTK 65 CKEN_SSP1
> > + CKEN_CAMERA CKEN_AC97CONF >;
> > + };
> > + };
> > +
> > };
>
> Maybe defining these clocks in pxa27x clock driver is better.
Agreed. After a long time trying to figure out the best DT binding for
clocks it seems that defining the per-clock data in DTS is not the best
way. The qcom msm and samsung exynos clock bindings and DTS data are
good examples.
Regards,
Mike
>
> Regards
> Haojian
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 0/4] Migrate PXA27x platforms to clock framework
2014-07-03 6:21 ` Haojian Zhuang
@ 2014-07-03 22:14 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-07-03 22:14 UTC (permalink / raw)
To: Haojian Zhuang
Cc: Arnd Bergmann, linux-arm-kernel@lists.infradead.org,
devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette, Eric Miao,
Mark Rutland
Haojian Zhuang <haojian.zhuang-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
>> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
>> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
>>
>
> It's acceptable to me that different silicons are queued in different stages.
> I only request that it won't break the compiler building & bootup.
OK.
>
> But I think that the pxa clock driver may be shared among all PXA silicons
> except for the clock table. What's your opinion?
I don't think so because of the core clocks.
These ones are specific to each pxa, and so their computation is :
- pxa25x plays with CCCR and specific L, M and N2 multiplication tables
- pxa27x plays with CCCR and specific L, M and N2 multiplication tables
- pxa3xx plays with ASCR, MEMCLKCFG, and AC97 div
I don't see very well how a clock table could describe that. Do you have
something specific in mind ?
>>> Also (for my understanding) when you say that you plan to do
>>> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
>> I don't have the Technical Reference Manuals for these ones so the answer is
>> no. And Google wasn't a great friend at providing them.
>>
>
> Converting them into new clock driver may not rely on the reference manual.
Actually I went through the code and :
- pxa26x is only a superset of pxa25x with one more clock : pxa26x-gpio
- pxa93x is the same clock set as pxa3xx, right ?
So if I convert pxa25x, pxa27x and pxa3xx I'll cover everything, right ?
>>> I assume it does as they are apparently minor revisions of the
>>> former, but it's not completely clear from your description.
>> My description doesn't mention them, as I have no information about them, nor
>> any hardware to test on.
>>
>
> We can request others to help testing in the mailing list.
That's exactly what will enable pxa25x and pxa3xx. I have them converted in my
tree, but I don't want to mix these patches in as I can't test them, and testing
brings in bigger delays.
Cheers.
--
Robert
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-07-03 22:14 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-07-03 22:14 UTC (permalink / raw)
To: linux-arm-kernel
Haojian Zhuang <haojian.zhuang@gmail.com> writes:
>> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
>> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
>>
>
> It's acceptable to me that different silicons are queued in different stages.
> I only request that it won't break the compiler building & bootup.
OK.
>
> But I think that the pxa clock driver may be shared among all PXA silicons
> except for the clock table. What's your opinion?
I don't think so because of the core clocks.
These ones are specific to each pxa, and so their computation is :
- pxa25x plays with CCCR and specific L, M and N2 multiplication tables
- pxa27x plays with CCCR and specific L, M and N2 multiplication tables
- pxa3xx plays with ASCR, MEMCLKCFG, and AC97 div
I don't see very well how a clock table could describe that. Do you have
something specific in mind ?
>>> Also (for my understanding) when you say that you plan to do
>>> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
>> I don't have the Technical Reference Manuals for these ones so the answer is
>> no. And Google wasn't a great friend at providing them.
>>
>
> Converting them into new clock driver may not rely on the reference manual.
Actually I went through the code and :
- pxa26x is only a superset of pxa25x with one more clock : pxa26x-gpio
- pxa93x is the same clock set as pxa3xx, right ?
So if I convert pxa25x, pxa27x and pxa3xx I'll cover everything, right ?
>>> I assume it does as they are apparently minor revisions of the
>>> former, but it's not completely clear from your description.
>> My description doesn't mention them, as I have no information about them, nor
>> any hardware to test on.
>>
>
> We can request others to help testing in the mailing list.
That's exactly what will enable pxa25x and pxa3xx. I have them converted in my
tree, but I don't want to mix these patches in as I can't test them, and testing
brings in bigger delays.
Cheers.
--
Robert
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 1/4] clk: add pxa27x clock drivers
2014-07-03 6:12 ` Haojian Zhuang
@ 2014-07-03 22:28 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-07-03 22:28 UTC (permalink / raw)
To: Haojian Zhuang
Cc: devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette, Eric Miao,
linux-arm-kernel@lists.infradead.org, Mark Rutland
Haojian Zhuang <haojian.zhuang-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
> On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org> wrote:
>> +static struct pxa27x_clocks_var_rate pxa27x_var_rate_clocks[] __initdata = {
>> + PXA2_VAR_RATE("pxa2xx-fb", true, LCD, &clk_pxa27x_lcd_ops),
>> + PXA2_VAR_RATE("pxa27x-camera.0", true, CAMERA, &clk_pxa27x_lcd_ops),
>> + PXA2_VAR_RATE("pxa2xx-pcmcia", false, MEMC, &clk_pxa27x_mem_ops),
>> +};
>> +
>
> It's not good to define a VAR table any more.
>
> Now we have the common clock framework. Since these three clocks are
> used as clock
> gate. And their clock frequency is controlled by other bits in
> peripheral controller registers.
I understand the clock hierarchy part.
> We can define parent & child clocks in clock tables. I think that you
> can get a lot of
> reference in clock drivers, such as exynos, hisilicon, ...
>
> Same comment on PXA_FIXED_RATE table.
That's a bit of work which will be thrown away anyway.
The goal is :
- keep same level of clock functionnality in non-DT builds
- have the full clock framework power for DT builds
- have non-DT builds removed once everything is ported to device-tree
Therefore, I don't want to redefine the full clock hierarchy in non-DT
builds. That's a lot of work for nothing. This patchset doesn't aim at a full
clock rework part, it aims at enabling porting all PXA platforms to DT without
breaking the non-DT part.
For DT, we already have this type of clock tree :
\u@\h:\w\$ cat /sys/kernel/debug/clk/clk_summary
clock enable_cnt prepare_cnt rate accuracy
--------------------------------------------------------------------------------
clk_dummy 1 1 0 0
im 0 0 0 0
osc32_768khz 1 1 32768 0
keypad 1 1 32768 0
osc13mhz 3 3 13000000 0
lcd 0 0 52000000 0
camera 0 0 52000000 0
memory 0 0 104000000 0
system bus 0 0 104000000 0
cpu core 0 0 104000000 0
halt-turbo mode 0 0 52000000 0
turbo mode 0 0 104000000 0
run mode 0 0 104000000 0
clk_ostimer 1 1 3250000 0
nssp,ostimer 1 1 3250000 0
clk_13mhz 2 2 13000000 0
ssp1 0 0 13000000 0
pwri2c 1 1 13000000 0
ssp3,hwuart 0 0 13000000 0
ssp2 0 0 13000000 0
pwm 1,3 0 0 13000000 0
pwm 0,2 1 1 13000000 0
pll_312mhz 3 3 312000000 0
clk_14_682mhz 0 0 12235294 0
i2s 0 0 12235294 0
clk_14_857mhz 1 3 14857142 0
btuart 1 2 14857142 0
ffuart 0 1 14857142 0
stuart 0 1 14857142 0
clk_19_5mhz 0 0 19500000 0
memstk 0 0 19500000 0
mmc 0 0 19500000 0
clk_32_842mhz 1 1 32842105 0
i2c 1 1 32842105 0
clk_48mhz 1 1 48000000 0
usim 0 0 48000000 0
msl 0 0 48000000 0
ficp 0 0 48000000 0
usb udc 1 1 48000000 0
usb host,assp 0 0 48000000 0
Cheers.
--
Robert
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 1/4] clk: add pxa27x clock drivers
@ 2014-07-03 22:28 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-07-03 22:28 UTC (permalink / raw)
To: linux-arm-kernel
Haojian Zhuang <haojian.zhuang@gmail.com> writes:
> On Mon, Jun 30, 2014 at 2:32 AM, Robert Jarzmik <robert.jarzmik@free.fr> wrote:
>> +static struct pxa27x_clocks_var_rate pxa27x_var_rate_clocks[] __initdata = {
>> + PXA2_VAR_RATE("pxa2xx-fb", true, LCD, &clk_pxa27x_lcd_ops),
>> + PXA2_VAR_RATE("pxa27x-camera.0", true, CAMERA, &clk_pxa27x_lcd_ops),
>> + PXA2_VAR_RATE("pxa2xx-pcmcia", false, MEMC, &clk_pxa27x_mem_ops),
>> +};
>> +
>
> It's not good to define a VAR table any more.
>
> Now we have the common clock framework. Since these three clocks are
> used as clock
> gate. And their clock frequency is controlled by other bits in
> peripheral controller registers.
I understand the clock hierarchy part.
> We can define parent & child clocks in clock tables. I think that you
> can get a lot of
> reference in clock drivers, such as exynos, hisilicon, ...
>
> Same comment on PXA_FIXED_RATE table.
That's a bit of work which will be thrown away anyway.
The goal is :
- keep same level of clock functionnality in non-DT builds
- have the full clock framework power for DT builds
- have non-DT builds removed once everything is ported to device-tree
Therefore, I don't want to redefine the full clock hierarchy in non-DT
builds. That's a lot of work for nothing. This patchset doesn't aim at a full
clock rework part, it aims at enabling porting all PXA platforms to DT without
breaking the non-DT part.
For DT, we already have this type of clock tree :
\u@\h:\w\$ cat /sys/kernel/debug/clk/clk_summary
clock enable_cnt prepare_cnt rate accuracy
--------------------------------------------------------------------------------
clk_dummy 1 1 0 0
im 0 0 0 0
osc32_768khz 1 1 32768 0
keypad 1 1 32768 0
osc13mhz 3 3 13000000 0
lcd 0 0 52000000 0
camera 0 0 52000000 0
memory 0 0 104000000 0
system bus 0 0 104000000 0
cpu core 0 0 104000000 0
halt-turbo mode 0 0 52000000 0
turbo mode 0 0 104000000 0
run mode 0 0 104000000 0
clk_ostimer 1 1 3250000 0
nssp,ostimer 1 1 3250000 0
clk_13mhz 2 2 13000000 0
ssp1 0 0 13000000 0
pwri2c 1 1 13000000 0
ssp3,hwuart 0 0 13000000 0
ssp2 0 0 13000000 0
pwm 1,3 0 0 13000000 0
pwm 0,2 1 1 13000000 0
pll_312mhz 3 3 312000000 0
clk_14_682mhz 0 0 12235294 0
i2s 0 0 12235294 0
clk_14_857mhz 1 3 14857142 0
btuart 1 2 14857142 0
ffuart 0 1 14857142 0
stuart 0 1 14857142 0
clk_19_5mhz 0 0 19500000 0
memstk 0 0 19500000 0
mmc 0 0 19500000 0
clk_32_842mhz 1 1 32842105 0
i2c 1 1 32842105 0
clk_48mhz 1 1 48000000 0
usim 0 0 48000000 0
msl 0 0 48000000 0
ficp 0 0 48000000 0
usb udc 1 1 48000000 0
usb host,assp 0 0 48000000 0
Cheers.
--
Robert
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 0/4] Migrate PXA27x platforms to clock framework
2014-07-03 22:14 ` Robert Jarzmik
@ 2014-07-04 2:39 ` Haojian Zhuang
-1 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-04 2:39 UTC (permalink / raw)
To: Robert Jarzmik
Cc: Mark Rutland, devicetree-u79uwXL29TY76Z2rM5mHXA, Mike Turquette,
Arnd Bergmann, Eric Miao,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r
On Fri, Jul 4, 2014 at 6:14 AM, Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org> wrote:
> Haojian Zhuang <haojian.zhuang-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
>
>>> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
>>> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
>>>
>>
>> It's acceptable to me that different silicons are queued in different stages.
>> I only request that it won't break the compiler building & bootup.
> OK.
>
>>
>> But I think that the pxa clock driver may be shared among all PXA silicons
>> except for the clock table. What's your opinion?
> I don't think so because of the core clocks.
> These ones are specific to each pxa, and so their computation is :
> - pxa25x plays with CCCR and specific L, M and N2 multiplication tables
> - pxa27x plays with CCCR and specific L, M and N2 multiplication tables
> - pxa3xx plays with ASCR, MEMCLKCFG, and AC97 div
>
> I don't see very well how a clock table could describe that. Do you have
> something specific in mind ?
>
As I remember, those registers bits are encoded. So we can define a clock mux
that connected to those fixed clock dividers. Then the clock gate is
the child of
the clock mux.
In DTS file, peripheral device node could specify a clock rate. In
peripheral device
driver, it could also change the clock rate by clk_set_rate().
>>>> Also (for my understanding) when you say that you plan to do
>>>> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
>>> I don't have the Technical Reference Manuals for these ones so the answer is
>>> no. And Google wasn't a great friend at providing them.
>>>
>>
>> Converting them into new clock driver may not rely on the reference manual.
> Actually I went through the code and :
> - pxa26x is only a superset of pxa25x with one more clock : pxa26x-gpio
> - pxa93x is the same clock set as pxa3xx, right ?
> So if I convert pxa25x, pxa27x and pxa3xx I'll cover everything, right ?
>
>>>> I assume it does as they are apparently minor revisions of the
>>>> former, but it's not completely clear from your description.
>>> My description doesn't mention them, as I have no information about them, nor
>>> any hardware to test on.
>>>
>>
>> We can request others to help testing in the mailing list.
> That's exactly what will enable pxa25x and pxa3xx. I have them converted in my
> tree, but I don't want to mix these patches in as I can't test them, and testing
> brings in bigger delays.
Delay shouldn't be the block issue.
Regards
Haojian
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 0/4] Migrate PXA27x platforms to clock framework
@ 2014-07-04 2:39 ` Haojian Zhuang
0 siblings, 0 replies; 32+ messages in thread
From: Haojian Zhuang @ 2014-07-04 2:39 UTC (permalink / raw)
To: linux-arm-kernel
On Fri, Jul 4, 2014 at 6:14 AM, Robert Jarzmik <robert.jarzmik@free.fr> wrote:
> Haojian Zhuang <haojian.zhuang@gmail.com> writes:
>
>>> Haojian, are you ok with that ? And BTW, does a combined kernel for PXA
>>> platforms even exists (mixing pxa3xx and pxa2xx for example) ?
>>>
>>
>> It's acceptable to me that different silicons are queued in different stages.
>> I only request that it won't break the compiler building & bootup.
> OK.
>
>>
>> But I think that the pxa clock driver may be shared among all PXA silicons
>> except for the clock table. What's your opinion?
> I don't think so because of the core clocks.
> These ones are specific to each pxa, and so their computation is :
> - pxa25x plays with CCCR and specific L, M and N2 multiplication tables
> - pxa27x plays with CCCR and specific L, M and N2 multiplication tables
> - pxa3xx plays with ASCR, MEMCLKCFG, and AC97 div
>
> I don't see very well how a clock table could describe that. Do you have
> something specific in mind ?
>
As I remember, those registers bits are encoded. So we can define a clock mux
that connected to those fixed clock dividers. Then the clock gate is
the child of
the clock mux.
In DTS file, peripheral device node could specify a clock rate. In
peripheral device
driver, it could also change the clock rate by clk_set_rate().
>>>> Also (for my understanding) when you say that you plan to do
>>>> pxa25x and pxa3xx next, does that include pxa26x and pxa93x?
>>> I don't have the Technical Reference Manuals for these ones so the answer is
>>> no. And Google wasn't a great friend at providing them.
>>>
>>
>> Converting them into new clock driver may not rely on the reference manual.
> Actually I went through the code and :
> - pxa26x is only a superset of pxa25x with one more clock : pxa26x-gpio
> - pxa93x is the same clock set as pxa3xx, right ?
> So if I convert pxa25x, pxa27x and pxa3xx I'll cover everything, right ?
>
>>>> I assume it does as they are apparently minor revisions of the
>>>> former, but it's not completely clear from your description.
>>> My description doesn't mention them, as I have no information about them, nor
>>> any hardware to test on.
>>>
>>
>> We can request others to help testing in the mailing list.
> That's exactly what will enable pxa25x and pxa3xx. I have them converted in my
> tree, but I don't want to mix these patches in as I can't test them, and testing
> brings in bigger delays.
Delay shouldn't be the block issue.
Regards
Haojian
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
2014-07-03 22:03 ` Mike Turquette
@ 2014-07-04 19:38 ` Robert Jarzmik
-1 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-07-04 19:38 UTC (permalink / raw)
To: Mike Turquette
Cc: Haojian Zhuang, devicetree-u79uwXL29TY76Z2rM5mHXA, Eric Miao,
linux-arm-kernel@lists.infradead.org, Mark Rutland
Mike Turquette <mturquette-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org> writes:
> Quoting Haojian Zhuang (2014-07-02 23:14:33)
>> Maybe defining these clocks in pxa27x clock driver is better.
>
> Agreed. After a long time trying to figure out the best DT binding for
> clocks it seems that defining the per-clock data in DTS is not the best
> way. The qcom msm and samsung exynos clock bindings and DTS data are
> good examples.
Ah, I see. Would have been a great comment in the former RFC patch serie, sic
... Maybe a comment in the clock device-tree documentation would prevent others
to do the same error ?
Well, it will take me some time to transform the dts data into code, especially
if I want to have some genericity to handle all PXA variants.
I'll work on it in the next days/week.
Cheers.
--
Robert
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 32+ messages in thread
* [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks
@ 2014-07-04 19:38 ` Robert Jarzmik
0 siblings, 0 replies; 32+ messages in thread
From: Robert Jarzmik @ 2014-07-04 19:38 UTC (permalink / raw)
To: linux-arm-kernel
Mike Turquette <mturquette@linaro.org> writes:
> Quoting Haojian Zhuang (2014-07-02 23:14:33)
>> Maybe defining these clocks in pxa27x clock driver is better.
>
> Agreed. After a long time trying to figure out the best DT binding for
> clocks it seems that defining the per-clock data in DTS is not the best
> way. The qcom msm and samsung exynos clock bindings and DTS data are
> good examples.
Ah, I see. Would have been a great comment in the former RFC patch serie, sic
... Maybe a comment in the clock device-tree documentation would prevent others
to do the same error ?
Well, it will take me some time to transform the dts data into code, especially
if I want to have some genericity to handle all PXA variants.
I'll work on it in the next days/week.
Cheers.
--
Robert
^ permalink raw reply [flat|nested] 32+ messages in thread
end of thread, other threads:[~2014-07-04 19:38 UTC | newest]
Thread overview: 32+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-06-29 18:32 [PATCH 0/4] Migrate PXA27x platforms to clock framework Robert Jarzmik
2014-06-29 18:32 ` Robert Jarzmik
[not found] ` <1404066744-13416-1-git-send-email-robert.jarzmik-GANU6spQydw@public.gmane.org>
2014-06-29 18:32 ` [PATCH 1/4] clk: add pxa27x clock drivers Robert Jarzmik
2014-06-29 18:32 ` Robert Jarzmik
[not found] ` <1404066744-13416-2-git-send-email-robert.jarzmik-GANU6spQydw@public.gmane.org>
2014-07-03 6:12 ` Haojian Zhuang
2014-07-03 6:12 ` Haojian Zhuang
[not found] ` <CAN1soZzoiGAz3OicdKbg5Dv2tW3yeinsu2i=M7KX+j1HuMs3Kg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-07-03 22:28 ` Robert Jarzmik
2014-07-03 22:28 ` Robert Jarzmik
2014-06-29 18:32 ` [PATCH 2/4] dts: add devicetree bindings for pxa27x clocks Robert Jarzmik
2014-06-29 18:32 ` Robert Jarzmik
[not found] ` <1404066744-13416-3-git-send-email-robert.jarzmik-GANU6spQydw@public.gmane.org>
2014-07-03 6:14 ` Haojian Zhuang
2014-07-03 6:14 ` Haojian Zhuang
[not found] ` <CAN1soZys4k6g5RgPLbreopOxTwDJ+bfPOq8XPwsNpyt5+YURjg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-07-03 22:03 ` Mike Turquette
2014-07-03 22:03 ` Mike Turquette
2014-07-04 19:38 ` Robert Jarzmik
2014-07-04 19:38 ` Robert Jarzmik
2014-06-29 18:32 ` [PATCH 3/4] arm: pxa: Transition pxa27x to clk framework Robert Jarzmik
2014-06-29 18:32 ` Robert Jarzmik
2014-06-29 18:32 ` [PATCH 4/4] clk: dts: document pxa27x clock binding Robert Jarzmik
2014-06-29 18:32 ` Robert Jarzmik
2014-06-30 6:55 ` [PATCH 0/4] Migrate PXA27x platforms to clock framework Arnd Bergmann
2014-06-30 6:55 ` Arnd Bergmann
2014-06-30 18:38 ` Robert Jarzmik
2014-06-30 18:38 ` Robert Jarzmik
[not found] ` <877g3yp7ie.fsf-GANU6spQydw@public.gmane.org>
2014-06-30 20:14 ` Arnd Bergmann
2014-06-30 20:14 ` Arnd Bergmann
2014-07-03 6:21 ` Haojian Zhuang
2014-07-03 6:21 ` Haojian Zhuang
[not found] ` <CAN1soZyu192Y7z-HJHo+3_bddFcf=H0udgJSWez1=Lmrb5wVfg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-07-03 22:14 ` Robert Jarzmik
2014-07-03 22:14 ` Robert Jarzmik
[not found] ` <87lhsanl7q.fsf-GANU6spQydw@public.gmane.org>
2014-07-04 2:39 ` Haojian Zhuang
2014-07-04 2:39 ` Haojian Zhuang
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.