* [RFC PATCHv3 0/6] arm: Initial TI-Nspire support
@ 2013-05-12 4:22 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel, linux
Cc: Daniel Tang, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
Changes between http://archive.arm.linux.org.uk/lurker/message/20130408.113343.585af217.en.html and v2:
* Added new drivers to support the irqchip and timers on older models.
* Added new device trees to support the other models.
Changes between v2 and v3:
* Clean up keypad driver
* Update copyright messages (2012->2013)
* Added clock driver
* Fix keypad support for classic models
* Fix nspire-classic-timer code to use updated CLOCKSOURCE_OF_DECLARE
* Change CLCD code to use panel capabilities
* Change UART clock to match APB speed
* Support for reset
First time using git send-email, please forgive me if this turns pear shaped.
Daniel Tang (6):
arm: Initial TI-Nspire support
arm: Add device trees for TI-Nspire
clk: Add TI-Nspire clock drivers
clocksource: Add TI-Nspire timer drivers
input: Add TI-Nspire keypad driver
irqchip: Add TI-Nspire irqchip
arch/arm/Kconfig | 2 +
arch/arm/Kconfig.debug | 16 ++
arch/arm/Makefile | 1 +
arch/arm/boot/dts/Makefile | 3 +
arch/arm/boot/dts/nspire-classic.dtsi | 75 +++++++
arch/arm/boot/dts/nspire-clp.dts | 45 +++++
arch/arm/boot/dts/nspire-cx.dts | 112 ++++++++++
arch/arm/boot/dts/nspire-tp.dts | 44 ++++
arch/arm/boot/dts/nspire.dtsi | 182 +++++++++++++++++
arch/arm/include/debug/nspire.S | 28 +++
arch/arm/mach-nspire/Kconfig | 15 ++
arch/arm/mach-nspire/Makefile | 2 +
arch/arm/mach-nspire/Makefile.boot | 0
arch/arm/mach-nspire/clcd.c | 117 +++++++++++
arch/arm/mach-nspire/clcd.h | 14 ++
arch/arm/mach-nspire/mmio.h | 18 ++
arch/arm/mach-nspire/nspire.c | 99 +++++++++
drivers/clk/Makefile | 1 +
drivers/clk/clk-nspire.c | 141 +++++++++++++
drivers/clocksource/Makefile | 1 +
drivers/clocksource/nspire-classic-timer.c | 199 ++++++++++++++++++
drivers/input/keyboard/Kconfig | 10 +
drivers/input/keyboard/Makefile | 1 +
drivers/input/keyboard/nspire-keypad.c | 315 +++++++++++++++++++++++++++++
drivers/irqchip/Makefile | 1 +
drivers/irqchip/irq-nspire-classic.c | 177 ++++++++++++++++
26 files changed, 1619 insertions(+)
create mode 100644 arch/arm/boot/dts/nspire-classic.dtsi
create mode 100644 arch/arm/boot/dts/nspire-clp.dts
create mode 100644 arch/arm/boot/dts/nspire-cx.dts
create mode 100644 arch/arm/boot/dts/nspire-tp.dts
create mode 100644 arch/arm/boot/dts/nspire.dtsi
create mode 100644 arch/arm/include/debug/nspire.S
create mode 100644 arch/arm/mach-nspire/Kconfig
create mode 100644 arch/arm/mach-nspire/Makefile
create mode 100644 arch/arm/mach-nspire/Makefile.boot
create mode 100644 arch/arm/mach-nspire/clcd.c
create mode 100644 arch/arm/mach-nspire/clcd.h
create mode 100644 arch/arm/mach-nspire/mmio.h
create mode 100644 arch/arm/mach-nspire/nspire.c
create mode 100644 drivers/clk/clk-nspire.c
create mode 100644 drivers/clocksource/nspire-classic-timer.c
create mode 100644 drivers/input/keyboard/nspire-keypad.c
create mode 100644 drivers/irqchip/irq-nspire-classic.c
--
1.8.1.3
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 0/6] arm: Initial TI-Nspire support
@ 2013-05-12 4:22 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel
Changes between http://archive.arm.linux.org.uk/lurker/message/20130408.113343.585af217.en.html and v2:
* Added new drivers to support the irqchip and timers on older models.
* Added new device trees to support the other models.
Changes between v2 and v3:
* Clean up keypad driver
* Update copyright messages (2012->2013)
* Added clock driver
* Fix keypad support for classic models
* Fix nspire-classic-timer code to use updated CLOCKSOURCE_OF_DECLARE
* Change CLCD code to use panel capabilities
* Change UART clock to match APB speed
* Support for reset
First time using git send-email, please forgive me if this turns pear shaped.
Daniel Tang (6):
arm: Initial TI-Nspire support
arm: Add device trees for TI-Nspire
clk: Add TI-Nspire clock drivers
clocksource: Add TI-Nspire timer drivers
input: Add TI-Nspire keypad driver
irqchip: Add TI-Nspire irqchip
arch/arm/Kconfig | 2 +
arch/arm/Kconfig.debug | 16 ++
arch/arm/Makefile | 1 +
arch/arm/boot/dts/Makefile | 3 +
arch/arm/boot/dts/nspire-classic.dtsi | 75 +++++++
arch/arm/boot/dts/nspire-clp.dts | 45 +++++
arch/arm/boot/dts/nspire-cx.dts | 112 ++++++++++
arch/arm/boot/dts/nspire-tp.dts | 44 ++++
arch/arm/boot/dts/nspire.dtsi | 182 +++++++++++++++++
arch/arm/include/debug/nspire.S | 28 +++
arch/arm/mach-nspire/Kconfig | 15 ++
arch/arm/mach-nspire/Makefile | 2 +
arch/arm/mach-nspire/Makefile.boot | 0
arch/arm/mach-nspire/clcd.c | 117 +++++++++++
arch/arm/mach-nspire/clcd.h | 14 ++
arch/arm/mach-nspire/mmio.h | 18 ++
arch/arm/mach-nspire/nspire.c | 99 +++++++++
drivers/clk/Makefile | 1 +
drivers/clk/clk-nspire.c | 141 +++++++++++++
drivers/clocksource/Makefile | 1 +
drivers/clocksource/nspire-classic-timer.c | 199 ++++++++++++++++++
drivers/input/keyboard/Kconfig | 10 +
drivers/input/keyboard/Makefile | 1 +
drivers/input/keyboard/nspire-keypad.c | 315 +++++++++++++++++++++++++++++
drivers/irqchip/Makefile | 1 +
drivers/irqchip/irq-nspire-classic.c | 177 ++++++++++++++++
26 files changed, 1619 insertions(+)
create mode 100644 arch/arm/boot/dts/nspire-classic.dtsi
create mode 100644 arch/arm/boot/dts/nspire-clp.dts
create mode 100644 arch/arm/boot/dts/nspire-cx.dts
create mode 100644 arch/arm/boot/dts/nspire-tp.dts
create mode 100644 arch/arm/boot/dts/nspire.dtsi
create mode 100644 arch/arm/include/debug/nspire.S
create mode 100644 arch/arm/mach-nspire/Kconfig
create mode 100644 arch/arm/mach-nspire/Makefile
create mode 100644 arch/arm/mach-nspire/Makefile.boot
create mode 100644 arch/arm/mach-nspire/clcd.c
create mode 100644 arch/arm/mach-nspire/clcd.h
create mode 100644 arch/arm/mach-nspire/mmio.h
create mode 100644 arch/arm/mach-nspire/nspire.c
create mode 100644 drivers/clk/clk-nspire.c
create mode 100644 drivers/clocksource/nspire-classic-timer.c
create mode 100644 drivers/input/keyboard/nspire-keypad.c
create mode 100644 drivers/irqchip/irq-nspire-classic.c
--
1.8.1.3
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 1/6] arm: Initial TI-Nspire support
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-12 4:22 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel, linux
Cc: Daniel Tang, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
arch/arm/Kconfig | 2 +
arch/arm/Kconfig.debug | 16 +++++
arch/arm/Makefile | 1 +
arch/arm/include/debug/nspire.S | 28 +++++++++
arch/arm/mach-nspire/Kconfig | 15 +++++
arch/arm/mach-nspire/Makefile | 2 +
arch/arm/mach-nspire/Makefile.boot | 0
arch/arm/mach-nspire/clcd.c | 117 +++++++++++++++++++++++++++++++++++++
arch/arm/mach-nspire/clcd.h | 14 +++++
arch/arm/mach-nspire/mmio.h | 18 ++++++
arch/arm/mach-nspire/nspire.c | 99 +++++++++++++++++++++++++++++++
11 files changed, 312 insertions(+)
create mode 100644 arch/arm/include/debug/nspire.S
create mode 100644 arch/arm/mach-nspire/Kconfig
create mode 100644 arch/arm/mach-nspire/Makefile
create mode 100644 arch/arm/mach-nspire/Makefile.boot
create mode 100644 arch/arm/mach-nspire/clcd.c
create mode 100644 arch/arm/mach-nspire/clcd.h
create mode 100644 arch/arm/mach-nspire/mmio.h
create mode 100644 arch/arm/mach-nspire/nspire.c
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index d423d58..b8cb225 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -963,6 +963,8 @@ source "arch/arm/mach-netx/Kconfig"
source "arch/arm/mach-nomadik/Kconfig"
+source "arch/arm/mach-nspire/Kconfig"
+
source "arch/arm/plat-omap/Kconfig"
source "arch/arm/mach-omap1/Kconfig"
diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug
index 1d41908..fe07941 100644
--- a/arch/arm/Kconfig.debug
+++ b/arch/arm/Kconfig.debug
@@ -309,6 +309,20 @@ choice
Say Y here if you want kernel low-level debugging support
on MVEBU based platforms.
+ config DEBUG_NSPIRE_CLASSIC_UART
+ bool "Kernel low-level debugging via TI-NSPIRE 8250 UART"
+ depends on ARCH_NSPIRE
+ help
+ Say Y here if you want kernel low-level debugging support
+ on TI-NSPIRE classic models.
+
+ config DEBUG_NSPIRE_CX_UART
+ bool "Kernel low-level debugging via TI-NSPIRE PL011 UART"
+ depends on ARCH_NSPIRE
+ help
+ Say Y here if you want kernel low-level debugging support
+ on TI-NSPIRE CX models.
+
config DEBUG_NOMADIK_UART
bool "Kernel low-level debugging messages via NOMADIK UART"
depends on ARCH_NOMADIK
@@ -633,6 +647,8 @@ config DEBUG_LL_INCLUDE
DEBUG_IMX53_UART ||\
DEBUG_IMX6Q_UART
default "debug/mvebu.S" if DEBUG_MVEBU_UART
+ default "debug/nspire.S" if DEBUG_NSPIRE_CX_UART || \
+ DEBUG_NSPIRE_CLASSIC_UART
default "debug/mxs.S" if DEBUG_IMX23_UART || DEBUG_IMX28_UART
default "debug/nomadik.S" if DEBUG_NOMADIK_UART
default "debug/omap2plus.S" if DEBUG_OMAP2PLUS_UART
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 4737408..a4340d1 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -164,6 +164,7 @@ machine-$(CONFIG_ARCH_MXS) += mxs
machine-$(CONFIG_ARCH_MVEBU) += mvebu
machine-$(CONFIG_ARCH_NETX) += netx
machine-$(CONFIG_ARCH_NOMADIK) += nomadik
+machine-$(CONFIG_ARCH_NSPIRE) += nspire
machine-$(CONFIG_ARCH_OMAP1) += omap1
machine-$(CONFIG_ARCH_OMAP2PLUS) += omap2
machine-$(CONFIG_ARCH_ORION5X) += orion5x
diff --git a/arch/arm/include/debug/nspire.S b/arch/arm/include/debug/nspire.S
new file mode 100644
index 0000000..886fd27
--- /dev/null
+++ b/arch/arm/include/debug/nspire.S
@@ -0,0 +1,28 @@
+/*
+ * linux/arch/arm/include/debug/nspire.S
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#define NSPIRE_EARLY_UART_PHYS_BASE 0x90020000
+#define NSPIRE_EARLY_UART_VIRT_BASE 0xfee20000
+
+.macro addruart, rp, rv, tmp
+ ldr \rp, =(NSPIRE_EARLY_UART_PHYS_BASE) @ physical base address
+ ldr \rv, =(NSPIRE_EARLY_UART_VIRT_BASE) @ virtual base address
+.endm
+
+
+#ifdef CONFIG_DEBUG_NSPIRE_CX_UART
+#include <asm/hardware/debug-pl01x.S>
+#endif
+
+#ifdef CONFIG_DEBUG_NSPIRE_CLASSIC_UART
+#define UART_SHIFT 2
+#include <asm/hardware/debug-8250.S>
+#endif
diff --git a/arch/arm/mach-nspire/Kconfig b/arch/arm/mach-nspire/Kconfig
new file mode 100644
index 0000000..a295b18
--- /dev/null
+++ b/arch/arm/mach-nspire/Kconfig
@@ -0,0 +1,15 @@
+config ARCH_NSPIRE
+ bool "TI-NSPIRE based"
+ depends on ARCH_MULTI_V4_V5
+ depends on MMU
+ select CPU_ARM926T
+ select COMMON_CLK
+ select GENERIC_CLOCKEVENTS
+ select SPARSE_IRQ
+ select ARM_AMBA
+ select ARM_VIC
+ select ARM_TIMER_SP804
+ select USE_OF
+ select CLKSRC_OF
+ help
+ This enables support for systems using the TI-NSPIRE CPU
diff --git a/arch/arm/mach-nspire/Makefile b/arch/arm/mach-nspire/Makefile
new file mode 100644
index 0000000..1bec256
--- /dev/null
+++ b/arch/arm/mach-nspire/Makefile
@@ -0,0 +1,2 @@
+obj-y += nspire.o
+obj-y += clcd.o
diff --git a/arch/arm/mach-nspire/Makefile.boot b/arch/arm/mach-nspire/Makefile.boot
new file mode 100644
index 0000000..e69de29
diff --git a/arch/arm/mach-nspire/clcd.c b/arch/arm/mach-nspire/clcd.c
new file mode 100644
index 0000000..a4b9f06
--- /dev/null
+++ b/arch/arm/mach-nspire/clcd.c
@@ -0,0 +1,117 @@
+/*
+ * linux/arch/arm/mach-nspire/clcd.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/of.h>
+#include <linux/amba/bus.h>
+#include <linux/amba/clcd.h>
+#include <linux/dma-mapping.h>
+
+static struct clcd_panel nspire_cx_lcd_panel = {
+ .mode = {
+ .name = "Color LCD",
+ .refresh = 60,
+ .xres = 320,
+ .yres = 240,
+ .sync = 0,
+ .vmode = FB_VMODE_NONINTERLACED,
+ .pixclock = 1,
+ .hsync_len = 6,
+ .vsync_len = 1,
+ .right_margin = 50,
+ .left_margin = 38,
+ .lower_margin = 3,
+ .upper_margin = 17,
+ },
+ .width = 65, /* ~6.50 cm */
+ .height = 49, /* ~4.87 cm */
+ .tim2 = TIM2_IPC,
+ .cntl = CNTL_LCDTFT | CNTL_LCDVCOMP(1),
+ .bpp = 16,
+};
+
+static struct clcd_panel nspire_classic_lcd_panel = {
+ .mode = {
+ .name = "Grayscale LCD",
+ .refresh = 60,
+ .xres = 320,
+ .yres = 240,
+ .sync = FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT,
+ .vmode = FB_VMODE_NONINTERLACED,
+ .pixclock = 1,
+ .hsync_len = 6,
+ .vsync_len = 1,
+ .right_margin = 6,
+ .left_margin = 6,
+ },
+ .width = 71, /* 7.11cm */
+ .height = 53, /* 5.33cm */
+ .tim2 = 0x80007d0,
+ .cntl = CNTL_LCDMONO8,
+ .bpp = 8,
+ .grayscale = 1
+};
+
+int nspire_clcd_setup(struct clcd_fb *fb)
+{
+ struct clcd_panel *panel;
+ size_t panel_size;
+ const char *type;
+ dma_addr_t dma;
+ int err;
+
+ BUG_ON(!fb->dev->dev.of_node);
+
+ err = of_property_read_string(fb->dev->dev.of_node, "lcd-type", &type);
+ if (err) {
+ pr_err("CLCD: Could not find lcd-type property\n");
+ return err;
+ }
+
+ if (!strcmp(type, "cx")) {
+ panel = &nspire_cx_lcd_panel;
+ } else if (!strcmp(type, "classic")) {
+ panel = &nspire_classic_lcd_panel;
+ } else {
+ pr_err("CLCD: Unknown lcd-type %s\n", type);
+ return -EINVAL;
+ }
+
+ panel_size = ((panel->mode.xres * panel->mode.yres) * panel->bpp) / 8;
+ panel_size = ALIGN(panel_size, PAGE_SIZE);
+
+ fb->fb.screen_base = dma_alloc_writecombine(&fb->dev->dev,
+ panel_size, &dma, GFP_KERNEL);
+
+ if (!fb->fb.screen_base) {
+ pr_err("CLCD: unable to map framebuffer\n");
+ return -ENOMEM;
+ }
+
+ fb->fb.fix.smem_start = dma;
+ fb->fb.fix.smem_len = panel_size;
+ fb->panel = panel;
+
+ return 0;
+}
+
+int nspire_clcd_mmap(struct clcd_fb *fb, struct vm_area_struct *vma)
+{
+ return dma_mmap_writecombine(&fb->dev->dev, vma,
+ fb->fb.screen_base, fb->fb.fix.smem_start,
+ fb->fb.fix.smem_len);
+}
+
+void nspire_clcd_remove(struct clcd_fb *fb)
+{
+ dma_free_writecombine(&fb->dev->dev, fb->fb.fix.smem_len,
+ fb->fb.screen_base, fb->fb.fix.smem_start);
+}
diff --git a/arch/arm/mach-nspire/clcd.h b/arch/arm/mach-nspire/clcd.h
new file mode 100644
index 0000000..8c33d2c
--- /dev/null
+++ b/arch/arm/mach-nspire/clcd.h
@@ -0,0 +1,14 @@
+/*
+ * linux/arch/arm/mach-nspire/clcd.h
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+int nspire_clcd_setup(struct clcd_fb *fb);
+int nspire_clcd_mmap(struct clcd_fb *fb, struct vm_area_struct *vma);
+void nspire_clcd_remove(struct clcd_fb *fb);
diff --git a/arch/arm/mach-nspire/mmio.h b/arch/arm/mach-nspire/mmio.h
new file mode 100644
index 0000000..9d45f59
--- /dev/null
+++ b/arch/arm/mach-nspire/mmio.h
@@ -0,0 +1,18 @@
+/*
+ * linux/arch/arm/mach-nspire/mmio.h
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#define NSPIRE_EARLY_UART_PHYS_BASE 0x90020000
+#define NSPIRE_EARLY_UART_VIRT_BASE 0xFEE20000
+
+#define NSPIRE_MISC_PHYS_BASE 0x900A0000
+#define NSPIRE_MISC_HWRESET 0x08
+
+#define NSPIRE_LCD_PHYS_BASE 0xC0000000
diff --git a/arch/arm/mach-nspire/nspire.c b/arch/arm/mach-nspire/nspire.c
new file mode 100644
index 0000000..821f3b2
--- /dev/null
+++ b/arch/arm/mach-nspire/nspire.c
@@ -0,0 +1,99 @@
+/*
+ * linux/arch/arm/mach-nspire/nspire.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/init.h>
+#include <linux/of_irq.h>
+#include <linux/of_address.h>
+#include <linux/of_platform.h>
+#include <linux/irqchip.h>
+#include <linux/irqchip/arm-vic.h>
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/amba/bus.h>
+#include <linux/amba/clcd.h>
+#include <linux/clocksource.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach-types.h>
+#include <asm/mach/map.h>
+
+#include <asm/hardware/timer-sp.h>
+
+#include "mmio.h"
+#include "clcd.h"
+
+static const char *nspire_dt_match[] __initconst = {
+ "arm,nspire",
+ "arm,nspire-cx",
+ "arm,nspire-tp",
+ "arm,nspire-clp",
+ NULL,
+};
+
+static struct map_desc nspire_io_desc[] __initdata = {
+ {
+ .virtual = NSPIRE_EARLY_UART_VIRT_BASE,
+ .pfn = __phys_to_pfn(NSPIRE_EARLY_UART_PHYS_BASE),
+ .length = SZ_4K,
+ .type = MT_DEVICE
+ }
+};
+
+static void __init nspire_map_io(void)
+{
+ iotable_init(nspire_io_desc, ARRAY_SIZE(nspire_io_desc));
+}
+
+static struct clcd_board nspire_clcd_data = {
+ .name = "LCD",
+ .caps = CLCD_CAP_ALL,
+ .check = clcdfb_check,
+ .decode = clcdfb_decode,
+ .setup = nspire_clcd_setup,
+ .mmap = nspire_clcd_mmap,
+ .remove = nspire_clcd_remove,
+};
+
+
+static struct of_dev_auxdata nspire_auxdata[] __initdata = {
+ OF_DEV_AUXDATA("arm,pl111", NSPIRE_LCD_PHYS_BASE,
+ NULL, &nspire_clcd_data),
+ { }
+};
+
+static void __init nspire_init(void)
+{
+ of_platform_populate(NULL, of_default_bus_match_table,
+ nspire_auxdata, NULL);
+}
+
+static void __init nspire_init_time(void)
+{
+ of_clk_init(NULL);
+ clocksource_of_init();
+}
+
+static void nspire_restart(char mode, const char *cmd)
+{
+ void __iomem *base = ioremap(NSPIRE_MISC_PHYS_BASE, SZ_4K);
+ if (!base)
+ return;
+
+ writel(2, base + NSPIRE_MISC_HWRESET);
+}
+
+DT_MACHINE_START(NSPIRE, "TI-NSPIRE")
+ .map_io = nspire_map_io,
+ .init_irq = irqchip_init,
+ .init_time = nspire_init_time,
+ .init_machine = nspire_init,
+ .dt_compat = nspire_dt_match,
+ .restart = nspire_restart,
+MACHINE_END
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 1/6] arm: Initial TI-Nspire support
@ 2013-05-12 4:22 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
arch/arm/Kconfig | 2 +
arch/arm/Kconfig.debug | 16 +++++
arch/arm/Makefile | 1 +
arch/arm/include/debug/nspire.S | 28 +++++++++
arch/arm/mach-nspire/Kconfig | 15 +++++
arch/arm/mach-nspire/Makefile | 2 +
arch/arm/mach-nspire/Makefile.boot | 0
arch/arm/mach-nspire/clcd.c | 117 +++++++++++++++++++++++++++++++++++++
arch/arm/mach-nspire/clcd.h | 14 +++++
arch/arm/mach-nspire/mmio.h | 18 ++++++
arch/arm/mach-nspire/nspire.c | 99 +++++++++++++++++++++++++++++++
11 files changed, 312 insertions(+)
create mode 100644 arch/arm/include/debug/nspire.S
create mode 100644 arch/arm/mach-nspire/Kconfig
create mode 100644 arch/arm/mach-nspire/Makefile
create mode 100644 arch/arm/mach-nspire/Makefile.boot
create mode 100644 arch/arm/mach-nspire/clcd.c
create mode 100644 arch/arm/mach-nspire/clcd.h
create mode 100644 arch/arm/mach-nspire/mmio.h
create mode 100644 arch/arm/mach-nspire/nspire.c
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index d423d58..b8cb225 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -963,6 +963,8 @@ source "arch/arm/mach-netx/Kconfig"
source "arch/arm/mach-nomadik/Kconfig"
+source "arch/arm/mach-nspire/Kconfig"
+
source "arch/arm/plat-omap/Kconfig"
source "arch/arm/mach-omap1/Kconfig"
diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug
index 1d41908..fe07941 100644
--- a/arch/arm/Kconfig.debug
+++ b/arch/arm/Kconfig.debug
@@ -309,6 +309,20 @@ choice
Say Y here if you want kernel low-level debugging support
on MVEBU based platforms.
+ config DEBUG_NSPIRE_CLASSIC_UART
+ bool "Kernel low-level debugging via TI-NSPIRE 8250 UART"
+ depends on ARCH_NSPIRE
+ help
+ Say Y here if you want kernel low-level debugging support
+ on TI-NSPIRE classic models.
+
+ config DEBUG_NSPIRE_CX_UART
+ bool "Kernel low-level debugging via TI-NSPIRE PL011 UART"
+ depends on ARCH_NSPIRE
+ help
+ Say Y here if you want kernel low-level debugging support
+ on TI-NSPIRE CX models.
+
config DEBUG_NOMADIK_UART
bool "Kernel low-level debugging messages via NOMADIK UART"
depends on ARCH_NOMADIK
@@ -633,6 +647,8 @@ config DEBUG_LL_INCLUDE
DEBUG_IMX53_UART ||\
DEBUG_IMX6Q_UART
default "debug/mvebu.S" if DEBUG_MVEBU_UART
+ default "debug/nspire.S" if DEBUG_NSPIRE_CX_UART || \
+ DEBUG_NSPIRE_CLASSIC_UART
default "debug/mxs.S" if DEBUG_IMX23_UART || DEBUG_IMX28_UART
default "debug/nomadik.S" if DEBUG_NOMADIK_UART
default "debug/omap2plus.S" if DEBUG_OMAP2PLUS_UART
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 4737408..a4340d1 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -164,6 +164,7 @@ machine-$(CONFIG_ARCH_MXS) += mxs
machine-$(CONFIG_ARCH_MVEBU) += mvebu
machine-$(CONFIG_ARCH_NETX) += netx
machine-$(CONFIG_ARCH_NOMADIK) += nomadik
+machine-$(CONFIG_ARCH_NSPIRE) += nspire
machine-$(CONFIG_ARCH_OMAP1) += omap1
machine-$(CONFIG_ARCH_OMAP2PLUS) += omap2
machine-$(CONFIG_ARCH_ORION5X) += orion5x
diff --git a/arch/arm/include/debug/nspire.S b/arch/arm/include/debug/nspire.S
new file mode 100644
index 0000000..886fd27
--- /dev/null
+++ b/arch/arm/include/debug/nspire.S
@@ -0,0 +1,28 @@
+/*
+ * linux/arch/arm/include/debug/nspire.S
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#define NSPIRE_EARLY_UART_PHYS_BASE 0x90020000
+#define NSPIRE_EARLY_UART_VIRT_BASE 0xfee20000
+
+.macro addruart, rp, rv, tmp
+ ldr \rp, =(NSPIRE_EARLY_UART_PHYS_BASE) @ physical base address
+ ldr \rv, =(NSPIRE_EARLY_UART_VIRT_BASE) @ virtual base address
+.endm
+
+
+#ifdef CONFIG_DEBUG_NSPIRE_CX_UART
+#include <asm/hardware/debug-pl01x.S>
+#endif
+
+#ifdef CONFIG_DEBUG_NSPIRE_CLASSIC_UART
+#define UART_SHIFT 2
+#include <asm/hardware/debug-8250.S>
+#endif
diff --git a/arch/arm/mach-nspire/Kconfig b/arch/arm/mach-nspire/Kconfig
new file mode 100644
index 0000000..a295b18
--- /dev/null
+++ b/arch/arm/mach-nspire/Kconfig
@@ -0,0 +1,15 @@
+config ARCH_NSPIRE
+ bool "TI-NSPIRE based"
+ depends on ARCH_MULTI_V4_V5
+ depends on MMU
+ select CPU_ARM926T
+ select COMMON_CLK
+ select GENERIC_CLOCKEVENTS
+ select SPARSE_IRQ
+ select ARM_AMBA
+ select ARM_VIC
+ select ARM_TIMER_SP804
+ select USE_OF
+ select CLKSRC_OF
+ help
+ This enables support for systems using the TI-NSPIRE CPU
diff --git a/arch/arm/mach-nspire/Makefile b/arch/arm/mach-nspire/Makefile
new file mode 100644
index 0000000..1bec256
--- /dev/null
+++ b/arch/arm/mach-nspire/Makefile
@@ -0,0 +1,2 @@
+obj-y += nspire.o
+obj-y += clcd.o
diff --git a/arch/arm/mach-nspire/Makefile.boot b/arch/arm/mach-nspire/Makefile.boot
new file mode 100644
index 0000000..e69de29
diff --git a/arch/arm/mach-nspire/clcd.c b/arch/arm/mach-nspire/clcd.c
new file mode 100644
index 0000000..a4b9f06
--- /dev/null
+++ b/arch/arm/mach-nspire/clcd.c
@@ -0,0 +1,117 @@
+/*
+ * linux/arch/arm/mach-nspire/clcd.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/of.h>
+#include <linux/amba/bus.h>
+#include <linux/amba/clcd.h>
+#include <linux/dma-mapping.h>
+
+static struct clcd_panel nspire_cx_lcd_panel = {
+ .mode = {
+ .name = "Color LCD",
+ .refresh = 60,
+ .xres = 320,
+ .yres = 240,
+ .sync = 0,
+ .vmode = FB_VMODE_NONINTERLACED,
+ .pixclock = 1,
+ .hsync_len = 6,
+ .vsync_len = 1,
+ .right_margin = 50,
+ .left_margin = 38,
+ .lower_margin = 3,
+ .upper_margin = 17,
+ },
+ .width = 65, /* ~6.50 cm */
+ .height = 49, /* ~4.87 cm */
+ .tim2 = TIM2_IPC,
+ .cntl = CNTL_LCDTFT | CNTL_LCDVCOMP(1),
+ .bpp = 16,
+};
+
+static struct clcd_panel nspire_classic_lcd_panel = {
+ .mode = {
+ .name = "Grayscale LCD",
+ .refresh = 60,
+ .xres = 320,
+ .yres = 240,
+ .sync = FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT,
+ .vmode = FB_VMODE_NONINTERLACED,
+ .pixclock = 1,
+ .hsync_len = 6,
+ .vsync_len = 1,
+ .right_margin = 6,
+ .left_margin = 6,
+ },
+ .width = 71, /* 7.11cm */
+ .height = 53, /* 5.33cm */
+ .tim2 = 0x80007d0,
+ .cntl = CNTL_LCDMONO8,
+ .bpp = 8,
+ .grayscale = 1
+};
+
+int nspire_clcd_setup(struct clcd_fb *fb)
+{
+ struct clcd_panel *panel;
+ size_t panel_size;
+ const char *type;
+ dma_addr_t dma;
+ int err;
+
+ BUG_ON(!fb->dev->dev.of_node);
+
+ err = of_property_read_string(fb->dev->dev.of_node, "lcd-type", &type);
+ if (err) {
+ pr_err("CLCD: Could not find lcd-type property\n");
+ return err;
+ }
+
+ if (!strcmp(type, "cx")) {
+ panel = &nspire_cx_lcd_panel;
+ } else if (!strcmp(type, "classic")) {
+ panel = &nspire_classic_lcd_panel;
+ } else {
+ pr_err("CLCD: Unknown lcd-type %s\n", type);
+ return -EINVAL;
+ }
+
+ panel_size = ((panel->mode.xres * panel->mode.yres) * panel->bpp) / 8;
+ panel_size = ALIGN(panel_size, PAGE_SIZE);
+
+ fb->fb.screen_base = dma_alloc_writecombine(&fb->dev->dev,
+ panel_size, &dma, GFP_KERNEL);
+
+ if (!fb->fb.screen_base) {
+ pr_err("CLCD: unable to map framebuffer\n");
+ return -ENOMEM;
+ }
+
+ fb->fb.fix.smem_start = dma;
+ fb->fb.fix.smem_len = panel_size;
+ fb->panel = panel;
+
+ return 0;
+}
+
+int nspire_clcd_mmap(struct clcd_fb *fb, struct vm_area_struct *vma)
+{
+ return dma_mmap_writecombine(&fb->dev->dev, vma,
+ fb->fb.screen_base, fb->fb.fix.smem_start,
+ fb->fb.fix.smem_len);
+}
+
+void nspire_clcd_remove(struct clcd_fb *fb)
+{
+ dma_free_writecombine(&fb->dev->dev, fb->fb.fix.smem_len,
+ fb->fb.screen_base, fb->fb.fix.smem_start);
+}
diff --git a/arch/arm/mach-nspire/clcd.h b/arch/arm/mach-nspire/clcd.h
new file mode 100644
index 0000000..8c33d2c
--- /dev/null
+++ b/arch/arm/mach-nspire/clcd.h
@@ -0,0 +1,14 @@
+/*
+ * linux/arch/arm/mach-nspire/clcd.h
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+int nspire_clcd_setup(struct clcd_fb *fb);
+int nspire_clcd_mmap(struct clcd_fb *fb, struct vm_area_struct *vma);
+void nspire_clcd_remove(struct clcd_fb *fb);
diff --git a/arch/arm/mach-nspire/mmio.h b/arch/arm/mach-nspire/mmio.h
new file mode 100644
index 0000000..9d45f59
--- /dev/null
+++ b/arch/arm/mach-nspire/mmio.h
@@ -0,0 +1,18 @@
+/*
+ * linux/arch/arm/mach-nspire/mmio.h
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#define NSPIRE_EARLY_UART_PHYS_BASE 0x90020000
+#define NSPIRE_EARLY_UART_VIRT_BASE 0xFEE20000
+
+#define NSPIRE_MISC_PHYS_BASE 0x900A0000
+#define NSPIRE_MISC_HWRESET 0x08
+
+#define NSPIRE_LCD_PHYS_BASE 0xC0000000
diff --git a/arch/arm/mach-nspire/nspire.c b/arch/arm/mach-nspire/nspire.c
new file mode 100644
index 0000000..821f3b2
--- /dev/null
+++ b/arch/arm/mach-nspire/nspire.c
@@ -0,0 +1,99 @@
+/*
+ * linux/arch/arm/mach-nspire/nspire.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/init.h>
+#include <linux/of_irq.h>
+#include <linux/of_address.h>
+#include <linux/of_platform.h>
+#include <linux/irqchip.h>
+#include <linux/irqchip/arm-vic.h>
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/amba/bus.h>
+#include <linux/amba/clcd.h>
+#include <linux/clocksource.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach-types.h>
+#include <asm/mach/map.h>
+
+#include <asm/hardware/timer-sp.h>
+
+#include "mmio.h"
+#include "clcd.h"
+
+static const char *nspire_dt_match[] __initconst = {
+ "arm,nspire",
+ "arm,nspire-cx",
+ "arm,nspire-tp",
+ "arm,nspire-clp",
+ NULL,
+};
+
+static struct map_desc nspire_io_desc[] __initdata = {
+ {
+ .virtual = NSPIRE_EARLY_UART_VIRT_BASE,
+ .pfn = __phys_to_pfn(NSPIRE_EARLY_UART_PHYS_BASE),
+ .length = SZ_4K,
+ .type = MT_DEVICE
+ }
+};
+
+static void __init nspire_map_io(void)
+{
+ iotable_init(nspire_io_desc, ARRAY_SIZE(nspire_io_desc));
+}
+
+static struct clcd_board nspire_clcd_data = {
+ .name = "LCD",
+ .caps = CLCD_CAP_ALL,
+ .check = clcdfb_check,
+ .decode = clcdfb_decode,
+ .setup = nspire_clcd_setup,
+ .mmap = nspire_clcd_mmap,
+ .remove = nspire_clcd_remove,
+};
+
+
+static struct of_dev_auxdata nspire_auxdata[] __initdata = {
+ OF_DEV_AUXDATA("arm,pl111", NSPIRE_LCD_PHYS_BASE,
+ NULL, &nspire_clcd_data),
+ { }
+};
+
+static void __init nspire_init(void)
+{
+ of_platform_populate(NULL, of_default_bus_match_table,
+ nspire_auxdata, NULL);
+}
+
+static void __init nspire_init_time(void)
+{
+ of_clk_init(NULL);
+ clocksource_of_init();
+}
+
+static void nspire_restart(char mode, const char *cmd)
+{
+ void __iomem *base = ioremap(NSPIRE_MISC_PHYS_BASE, SZ_4K);
+ if (!base)
+ return;
+
+ writel(2, base + NSPIRE_MISC_HWRESET);
+}
+
+DT_MACHINE_START(NSPIRE, "TI-NSPIRE")
+ .map_io = nspire_map_io,
+ .init_irq = irqchip_init,
+ .init_time = nspire_init_time,
+ .init_machine = nspire_init,
+ .dt_compat = nspire_dt_match,
+ .restart = nspire_restart,
+MACHINE_END
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 2/6] arm: Add device trees for TI-Nspire
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-12 4:22 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel, linux
Cc: Daniel Tang, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
arch/arm/boot/dts/Makefile | 3 +
arch/arm/boot/dts/nspire-classic.dtsi | 75 ++++++++++++++
arch/arm/boot/dts/nspire-clp.dts | 45 +++++++++
arch/arm/boot/dts/nspire-cx.dts | 112 +++++++++++++++++++++
arch/arm/boot/dts/nspire-tp.dts | 44 ++++++++
arch/arm/boot/dts/nspire.dtsi | 182 ++++++++++++++++++++++++++++++++++
6 files changed, 461 insertions(+)
create mode 100644 arch/arm/boot/dts/nspire-classic.dtsi
create mode 100644 arch/arm/boot/dts/nspire-clp.dts
create mode 100644 arch/arm/boot/dts/nspire-cx.dts
create mode 100644 arch/arm/boot/dts/nspire-tp.dts
create mode 100644 arch/arm/boot/dts/nspire.dtsi
diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
index b9f7121..5677d6c 100644
--- a/arch/arm/boot/dts/Makefile
+++ b/arch/arm/boot/dts/Makefile
@@ -135,6 +135,9 @@ dtb-$(CONFIG_ARCH_MXS) += imx23-evk.dtb \
imx28-sps1.dtb \
imx28-tx28.dtb
dtb-$(CONFIG_ARCH_NOMADIK) += ste-nomadik-s8815.dtb
+dtb-$(CONFIG_ARCH_NSPIRE) += nspire-cx.dtb \
+ nspire-tp.dtb \
+ nspire-clp.dtb
dtb-$(CONFIG_ARCH_OMAP2PLUS) += omap2420-h4.dtb \
omap3430-sdp.dtb \
omap3-beagle.dtb \
diff --git a/arch/arm/boot/dts/nspire-classic.dtsi b/arch/arm/boot/dts/nspire-classic.dtsi
new file mode 100644
index 0000000..eb1c2a2
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-classic.dtsi
@@ -0,0 +1,75 @@
+/*
+ * linux/arch/arm/boot/nspire-classic.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+/include/ "nspire.dtsi"
+
+&lcd {
+ lcd-type = "classic";
+};
+
+&fast_timer {
+ /* compatible = "nspire-classic-timer"; */
+ reg = <0x90010000 0x1000>, <0x900A0010 0x8>;
+};
+
+&uart {
+ compatible = "ns16550";
+ reg-shift = <2>;
+ reg-io-width = <4>;
+ clocks = <&apb_pclk>;
+ clock-names = "apb_pclk";
+ no-loopback-test;
+};
+
+&timer0 {
+ /* compatible = "nspire-classic-timer"; */
+ reg = <0x900C0000 0x1000>, <0x900A0018 0x8>;
+};
+
+&timer1 {
+ compatible = "nspire-classic-timer";
+ reg = <0x900D0000 0x1000>, <0x900A0020 0x8>;
+};
+
+&keypad {
+ active-low;
+
+};
+
+&base_clk {
+ io-type = "classic";
+};
+
+&ahb_clk {
+ io-type = "classic";
+};
+
+/ {
+ memory {
+ device_type = "memory";
+ reg = <0x10000000 0x2000000>; /* 32 MB */
+ };
+
+ ahb {
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ intc: interrupt-controller@DC000000 {
+ compatible = "nspire-classic-intc";
+ interrupt-controller;
+ reg = <0xDC000000 0x1000>;
+ #interrupt-cells = <1>;
+ };
+ };
+ chosen {
+ bootargs = "debug earlyprintk console=tty0 console=ttyS0,115200n8 root=/dev/ram0";
+ };
+};
diff --git a/arch/arm/boot/dts/nspire-clp.dts b/arch/arm/boot/dts/nspire-clp.dts
new file mode 100644
index 0000000..7f25f44
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-clp.dts
@@ -0,0 +1,45 @@
+/*
+ * linux/arch/arm/boot/nspire-clp.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+/dts-v1/;
+
+/include/ "nspire-classic.dtsi"
+
+&keypad {
+ keymap = <
+ 0x0000001c 0x0001001c 0x00020039
+ 0x0004002c 0x00050034 0x00060015
+ 0x0007000b 0x0008002d 0x01000033
+ 0x0101004e 0x01020011 0x01030004
+ 0x0104002f 0x01050003 0x01060016
+ 0x01070002 0x01080014 0x02000062
+ 0x0201000c 0x0202001f 0x02030007
+ 0x02040013 0x02050006 0x02060010
+ 0x02070005 0x02080019 0x03000027
+ 0x03010037 0x03020018 0x0303000a
+ 0x03040031 0x03050009 0x03060032
+ 0x03070008 0x03080026 0x04000028
+ 0x04010035 0x04020025 0x04040024
+ 0x04060017 0x04080023 0x05000028
+ 0x05020022 0x0503001b 0x05040021
+ 0x0505001a 0x05060012 0x0507006f
+ 0x05080020 0x0509002a 0x0601001c
+ 0x0602002e 0x06030068 0x06040030
+ 0x0605006d 0x0606001e 0x06070001
+ 0x0608002b 0x0609000f 0x07000067
+ 0x0702006a 0x0704006c 0x07060069
+ 0x0707000e 0x0708001d 0x070a000d
+ >;
+};
+
+/ {
+ model = "TI-NSPIRE Clickpad";
+ compatible = "arm,nspire-clp";
+};
diff --git a/arch/arm/boot/dts/nspire-cx.dts b/arch/arm/boot/dts/nspire-cx.dts
new file mode 100644
index 0000000..ac5c2ad
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-cx.dts
@@ -0,0 +1,112 @@
+/*
+ * linux/arch/arm/boot/nspire-cx.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+/dts-v1/;
+
+/include/ "nspire.dtsi"
+
+&lcd {
+ lcd-type = "cx";
+};
+
+&fast_timer {
+ /* compatible = "arm,sp804", "arm,primecell"; */
+};
+
+&uart {
+ compatible = "arm,pl011", "arm,primecell";
+
+ clocks = <&uart_clk>, <&apb_pclk>;
+ clock-names = "uart_clk", "apb_pclk";
+};
+
+&timer0 {
+ compatible = "arm,sp804", "arm,primecell";
+};
+
+&timer1 {
+ compatible = "arm,sp804", "arm,primecell";
+};
+
+&base_clk {
+ io-type = "cx";
+};
+
+&ahb_clk {
+ io-type = "cx";
+};
+
+&keypad {
+ keymap = <
+ 0x0000001c 0x0001001c 0x00040039
+ 0x0005002c 0x00060015 0x0007000b
+ 0x0008000f 0x0100002d 0x01010011
+ 0x0102002f 0x01030004 0x01040016
+ 0x01050014 0x0106001f 0x01070002
+ 0x010a006a 0x02000013 0x02010010
+ 0x02020019 0x02030007 0x02040018
+ 0x02050031 0x02060032 0x02070005
+ 0x02080028 0x0209006c 0x03000026
+ 0x03010025 0x03020024 0x0303000a
+ 0x03040017 0x03050023 0x03060022
+ 0x03070008 0x03080035 0x03090069
+ 0x04000021 0x04010012 0x04020020
+ 0x0404002e 0x04050030 0x0406001e
+ 0x0407000d 0x04080037 0x04090067
+ 0x05010038 0x0502000c 0x0503001b
+ 0x05040034 0x0505001a 0x05060006
+ 0x05080027 0x0509000e 0x050a006f
+ 0x0600002b 0x0602004e 0x06030068
+ 0x06040003 0x0605006d 0x06060009
+ 0x06070001 0x0609000f 0x0708002a
+ 0x0709001d 0x070a0033 >;
+};
+
+/ {
+ model = "TI-NSPIRE CX";
+ compatible = "arm,nspire-cx";
+
+ memory {
+ device_type = "memory";
+ reg = <0x10000000 0x4000000>; /* 64 MB */
+ };
+
+ uart_clk: uart_clk {
+ #clock-cells = <0>;
+ compatible = "fixed-clock";
+ clock-frequency = <12000000>;
+ };
+
+ ahb {
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ intc: interrupt-controller@DC000000 {
+ compatible = "arm,pl190-vic";
+ interrupt-controller;
+ reg = <0xDC000000 0x1000>;
+ #interrupt-cells = <1>;
+ };
+
+ apb@90000000 {
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ i2c@90050000 {
+ compatible = "snps,designware-i2c";
+ reg = <0x90050000 0x1000>;
+ interrupts = <20>;
+ };
+ };
+ };
+ chosen {
+ bootargs = "debug earlyprintk console=tty0 console=ttyAMA0,115200n8 root=/dev/ram0";
+ };
+};
diff --git a/arch/arm/boot/dts/nspire-tp.dts b/arch/arm/boot/dts/nspire-tp.dts
new file mode 100644
index 0000000..485d0ff
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-tp.dts
@@ -0,0 +1,44 @@
+/*
+ * linux/arch/arm/boot/nspire-tp.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+/dts-v1/;
+
+/include/ "nspire-classic.dtsi"
+
+&keypad {
+ keymap = <
+ 0x0000001c 0x0001001c 0x00040039
+ 0x0005002c 0x00060015 0x0007000b
+ 0x0008000f 0x0100002d 0x01010011
+ 0x0102002f 0x01030004 0x01040016
+ 0x01050014 0x0106001f 0x01070002
+ 0x010a006a 0x02000013 0x02010010
+ 0x02020019 0x02030007 0x02040018
+ 0x02050031 0x02060032 0x02070005
+ 0x02080028 0x0209006c 0x03000026
+ 0x03010025 0x03020024 0x0303000a
+ 0x03040017 0x03050023 0x03060022
+ 0x03070008 0x03080035 0x03090069
+ 0x04000021 0x04010012 0x04020020
+ 0x0404002e 0x04050030 0x0406001e
+ 0x0407000d 0x04080037 0x04090067
+ 0x05010038 0x0502000c 0x0503001b
+ 0x05040034 0x0505001a 0x05060006
+ 0x05080027 0x0509000e 0x050a006f
+ 0x0600002b 0x0602004e 0x06030068
+ 0x06040003 0x0605006d 0x06060009
+ 0x06070001 0x0609000f 0x0708002a
+ 0x0709001d 0x070a0033 >;
+};
+
+/ {
+ model = "TI-NSPIRE Touchpad";
+ compatible = "arm,nspire-tp";
+};
diff --git a/arch/arm/boot/dts/nspire.dtsi b/arch/arm/boot/dts/nspire.dtsi
new file mode 100644
index 0000000..0e780f9
--- /dev/null
+++ b/arch/arm/boot/dts/nspire.dtsi
@@ -0,0 +1,182 @@
+/*
+ * linux/arch/arm/boot/nspire.dtsi
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+/include/ "skeleton.dtsi"
+
+/ {
+ interrupt-parent = <&intc>;
+
+ cpus {
+ cpu@0 {
+ compatible = "arm,arm926ejs";
+ };
+ };
+
+ bootrom: bootrom@00000000 {
+ reg = <0x00000000 0x80000>;
+ };
+
+ sram: sram@A4000000 {
+ device = "memory";
+ reg = <0xA4000000 0x20000>;
+ };
+
+ timer_clk: timer_clk {
+ #clock-cells = <0>;
+ compatible = "fixed-clock";
+ clock-frequency = <32768>;
+ };
+
+ base_clk: base_clk {
+ #clock-cells = <0>;
+ compatible = "nspire-clock";
+ reg = <0x900B0000 0x2>;
+ };
+
+ ahb_clk: ahb_clk {
+ #clock-cells = <0>;
+ compatible = "nspire-ahb-divider";
+ reg = <0x900B0000 0x2>;
+ clocks = <&base_clk>;
+ clock-names = "base_clk";
+ };
+
+ apb_pclk: apb_pclk {
+ #clock-cells = <0>;
+ compatible = "fixed-factor-clock";
+ clock-div = <2>;
+ clock-mult = <1>;
+ clocks = <&ahb_clk>;
+ clock-names = "ahb_clk";
+ };
+
+ ahb {
+ compatible = "simple-bus";
+ #address-cells = <1>;
+ #size-cells = <1>;
+ ranges;
+
+ spi: spi@A9000000 {
+ reg = <0xA9000000 0x1000>;
+ };
+
+ usb0: usb@B0000000 {
+ reg = <0xB0000000 0x1000>;
+ interrupts = <8>;
+ };
+
+ usb1: usb@B4000000 {
+ reg = <0xB4000000 0x1000>;
+ interrupts = <9>;
+ status = "disabled";
+ };
+
+ lcd: lcd@C0000000 {
+ compatible = "arm,pl111", "arm,primecell";
+ reg = <0xC0000000 0x1000>;
+ interrupts = <21>;
+
+ clocks = <&apb_pclk>;
+ clock-names = "apb_pclk";
+ };
+
+ adc: adc@C4000000 {
+ reg = <0xC4000000 0x1000>;
+ interrupts = <11>;
+ };
+
+ tdes: crypto@C8010000 {
+ reg = <0xC8010000 0x1000>;
+ };
+
+ sha256: crypto@CC000000 {
+ reg = <0xCC000000 0x1000>;
+ };
+
+ apb@90000000 {
+ compatible = "simple-bus";
+ #address-cells = <1>;
+ #size-cells = <1>;
+ clock-ranges;
+ ranges;
+
+ gpio: gpio@90000000 {
+ reg = <0x90000000 0x1000>;
+ interrupts = <7>;
+ };
+
+ fast_timer: timer@90010000 {
+ reg = <0x90010000 0x1000>;
+ interrupts = <17>;
+ };
+
+ uart: serial@90020000 {
+ reg = <0x90020000 0x1000>;
+ interrupts = <1>;
+ };
+
+ timer0: timer@900C0000 {
+ reg = <0x900C0000 0x1000>;
+
+ clocks = <&timer_clk>;
+ clock-names = "timer_clk";
+ };
+
+ timer1: timer@900D0000 {
+ reg = <0x900D0000 0x1000>;
+ interrupts = <19>;
+
+ clocks = <&timer_clk>;
+ clock-names = "timer_clk";
+ };
+
+ watchdog: watchdog@90060000 {
+ compatible = "arm,amba-primecell";
+ reg = <0x90060000 0x1000>;
+ interrupts = <3>;
+ };
+
+ rtc: rtc@90090000 {
+ reg = <0x90090000 0x1000>;
+ interrupts = <4>;
+ };
+
+ misc: misc@900A0000 {
+ reg = <0x900A0000 0x1000>;
+ };
+
+ pwr: pwr@900B0000 {
+ reg = <0x900B0000 0x1000>;
+ interrupts = <15>;
+ };
+
+ keypad: input@900E0000 {
+ compatible = "nspire-keypad";
+ reg = <0x900E0000 0x1000>;
+ interrupts = <16>;
+
+ scan-interval = <1000>;
+ row-delay = <200>;
+
+ clocks = <&apb_pclk>;
+ clock-names = "apb_pclk";
+ };
+
+ contrast: contrast@900F0000 {
+ reg = <0x900F0000 0x1000>;
+ };
+
+ led: led@90110000 {
+ reg = <0x90110000 0x1000>;
+ };
+ };
+ };
+};
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 2/6] arm: Add device trees for TI-Nspire
@ 2013-05-12 4:22 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
arch/arm/boot/dts/Makefile | 3 +
arch/arm/boot/dts/nspire-classic.dtsi | 75 ++++++++++++++
arch/arm/boot/dts/nspire-clp.dts | 45 +++++++++
arch/arm/boot/dts/nspire-cx.dts | 112 +++++++++++++++++++++
arch/arm/boot/dts/nspire-tp.dts | 44 ++++++++
arch/arm/boot/dts/nspire.dtsi | 182 ++++++++++++++++++++++++++++++++++
6 files changed, 461 insertions(+)
create mode 100644 arch/arm/boot/dts/nspire-classic.dtsi
create mode 100644 arch/arm/boot/dts/nspire-clp.dts
create mode 100644 arch/arm/boot/dts/nspire-cx.dts
create mode 100644 arch/arm/boot/dts/nspire-tp.dts
create mode 100644 arch/arm/boot/dts/nspire.dtsi
diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
index b9f7121..5677d6c 100644
--- a/arch/arm/boot/dts/Makefile
+++ b/arch/arm/boot/dts/Makefile
@@ -135,6 +135,9 @@ dtb-$(CONFIG_ARCH_MXS) += imx23-evk.dtb \
imx28-sps1.dtb \
imx28-tx28.dtb
dtb-$(CONFIG_ARCH_NOMADIK) += ste-nomadik-s8815.dtb
+dtb-$(CONFIG_ARCH_NSPIRE) += nspire-cx.dtb \
+ nspire-tp.dtb \
+ nspire-clp.dtb
dtb-$(CONFIG_ARCH_OMAP2PLUS) += omap2420-h4.dtb \
omap3430-sdp.dtb \
omap3-beagle.dtb \
diff --git a/arch/arm/boot/dts/nspire-classic.dtsi b/arch/arm/boot/dts/nspire-classic.dtsi
new file mode 100644
index 0000000..eb1c2a2
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-classic.dtsi
@@ -0,0 +1,75 @@
+/*
+ * linux/arch/arm/boot/nspire-classic.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+/include/ "nspire.dtsi"
+
+&lcd {
+ lcd-type = "classic";
+};
+
+&fast_timer {
+ /* compatible = "nspire-classic-timer"; */
+ reg = <0x90010000 0x1000>, <0x900A0010 0x8>;
+};
+
+&uart {
+ compatible = "ns16550";
+ reg-shift = <2>;
+ reg-io-width = <4>;
+ clocks = <&apb_pclk>;
+ clock-names = "apb_pclk";
+ no-loopback-test;
+};
+
+&timer0 {
+ /* compatible = "nspire-classic-timer"; */
+ reg = <0x900C0000 0x1000>, <0x900A0018 0x8>;
+};
+
+&timer1 {
+ compatible = "nspire-classic-timer";
+ reg = <0x900D0000 0x1000>, <0x900A0020 0x8>;
+};
+
+&keypad {
+ active-low;
+
+};
+
+&base_clk {
+ io-type = "classic";
+};
+
+&ahb_clk {
+ io-type = "classic";
+};
+
+/ {
+ memory {
+ device_type = "memory";
+ reg = <0x10000000 0x2000000>; /* 32 MB */
+ };
+
+ ahb {
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ intc: interrupt-controller at DC000000 {
+ compatible = "nspire-classic-intc";
+ interrupt-controller;
+ reg = <0xDC000000 0x1000>;
+ #interrupt-cells = <1>;
+ };
+ };
+ chosen {
+ bootargs = "debug earlyprintk console=tty0 console=ttyS0,115200n8 root=/dev/ram0";
+ };
+};
diff --git a/arch/arm/boot/dts/nspire-clp.dts b/arch/arm/boot/dts/nspire-clp.dts
new file mode 100644
index 0000000..7f25f44
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-clp.dts
@@ -0,0 +1,45 @@
+/*
+ * linux/arch/arm/boot/nspire-clp.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+/dts-v1/;
+
+/include/ "nspire-classic.dtsi"
+
+&keypad {
+ keymap = <
+ 0x0000001c 0x0001001c 0x00020039
+ 0x0004002c 0x00050034 0x00060015
+ 0x0007000b 0x0008002d 0x01000033
+ 0x0101004e 0x01020011 0x01030004
+ 0x0104002f 0x01050003 0x01060016
+ 0x01070002 0x01080014 0x02000062
+ 0x0201000c 0x0202001f 0x02030007
+ 0x02040013 0x02050006 0x02060010
+ 0x02070005 0x02080019 0x03000027
+ 0x03010037 0x03020018 0x0303000a
+ 0x03040031 0x03050009 0x03060032
+ 0x03070008 0x03080026 0x04000028
+ 0x04010035 0x04020025 0x04040024
+ 0x04060017 0x04080023 0x05000028
+ 0x05020022 0x0503001b 0x05040021
+ 0x0505001a 0x05060012 0x0507006f
+ 0x05080020 0x0509002a 0x0601001c
+ 0x0602002e 0x06030068 0x06040030
+ 0x0605006d 0x0606001e 0x06070001
+ 0x0608002b 0x0609000f 0x07000067
+ 0x0702006a 0x0704006c 0x07060069
+ 0x0707000e 0x0708001d 0x070a000d
+ >;
+};
+
+/ {
+ model = "TI-NSPIRE Clickpad";
+ compatible = "arm,nspire-clp";
+};
diff --git a/arch/arm/boot/dts/nspire-cx.dts b/arch/arm/boot/dts/nspire-cx.dts
new file mode 100644
index 0000000..ac5c2ad
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-cx.dts
@@ -0,0 +1,112 @@
+/*
+ * linux/arch/arm/boot/nspire-cx.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+/dts-v1/;
+
+/include/ "nspire.dtsi"
+
+&lcd {
+ lcd-type = "cx";
+};
+
+&fast_timer {
+ /* compatible = "arm,sp804", "arm,primecell"; */
+};
+
+&uart {
+ compatible = "arm,pl011", "arm,primecell";
+
+ clocks = <&uart_clk>, <&apb_pclk>;
+ clock-names = "uart_clk", "apb_pclk";
+};
+
+&timer0 {
+ compatible = "arm,sp804", "arm,primecell";
+};
+
+&timer1 {
+ compatible = "arm,sp804", "arm,primecell";
+};
+
+&base_clk {
+ io-type = "cx";
+};
+
+&ahb_clk {
+ io-type = "cx";
+};
+
+&keypad {
+ keymap = <
+ 0x0000001c 0x0001001c 0x00040039
+ 0x0005002c 0x00060015 0x0007000b
+ 0x0008000f 0x0100002d 0x01010011
+ 0x0102002f 0x01030004 0x01040016
+ 0x01050014 0x0106001f 0x01070002
+ 0x010a006a 0x02000013 0x02010010
+ 0x02020019 0x02030007 0x02040018
+ 0x02050031 0x02060032 0x02070005
+ 0x02080028 0x0209006c 0x03000026
+ 0x03010025 0x03020024 0x0303000a
+ 0x03040017 0x03050023 0x03060022
+ 0x03070008 0x03080035 0x03090069
+ 0x04000021 0x04010012 0x04020020
+ 0x0404002e 0x04050030 0x0406001e
+ 0x0407000d 0x04080037 0x04090067
+ 0x05010038 0x0502000c 0x0503001b
+ 0x05040034 0x0505001a 0x05060006
+ 0x05080027 0x0509000e 0x050a006f
+ 0x0600002b 0x0602004e 0x06030068
+ 0x06040003 0x0605006d 0x06060009
+ 0x06070001 0x0609000f 0x0708002a
+ 0x0709001d 0x070a0033 >;
+};
+
+/ {
+ model = "TI-NSPIRE CX";
+ compatible = "arm,nspire-cx";
+
+ memory {
+ device_type = "memory";
+ reg = <0x10000000 0x4000000>; /* 64 MB */
+ };
+
+ uart_clk: uart_clk {
+ #clock-cells = <0>;
+ compatible = "fixed-clock";
+ clock-frequency = <12000000>;
+ };
+
+ ahb {
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ intc: interrupt-controller at DC000000 {
+ compatible = "arm,pl190-vic";
+ interrupt-controller;
+ reg = <0xDC000000 0x1000>;
+ #interrupt-cells = <1>;
+ };
+
+ apb at 90000000 {
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ i2c at 90050000 {
+ compatible = "snps,designware-i2c";
+ reg = <0x90050000 0x1000>;
+ interrupts = <20>;
+ };
+ };
+ };
+ chosen {
+ bootargs = "debug earlyprintk console=tty0 console=ttyAMA0,115200n8 root=/dev/ram0";
+ };
+};
diff --git a/arch/arm/boot/dts/nspire-tp.dts b/arch/arm/boot/dts/nspire-tp.dts
new file mode 100644
index 0000000..485d0ff
--- /dev/null
+++ b/arch/arm/boot/dts/nspire-tp.dts
@@ -0,0 +1,44 @@
+/*
+ * linux/arch/arm/boot/nspire-tp.dts
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+/dts-v1/;
+
+/include/ "nspire-classic.dtsi"
+
+&keypad {
+ keymap = <
+ 0x0000001c 0x0001001c 0x00040039
+ 0x0005002c 0x00060015 0x0007000b
+ 0x0008000f 0x0100002d 0x01010011
+ 0x0102002f 0x01030004 0x01040016
+ 0x01050014 0x0106001f 0x01070002
+ 0x010a006a 0x02000013 0x02010010
+ 0x02020019 0x02030007 0x02040018
+ 0x02050031 0x02060032 0x02070005
+ 0x02080028 0x0209006c 0x03000026
+ 0x03010025 0x03020024 0x0303000a
+ 0x03040017 0x03050023 0x03060022
+ 0x03070008 0x03080035 0x03090069
+ 0x04000021 0x04010012 0x04020020
+ 0x0404002e 0x04050030 0x0406001e
+ 0x0407000d 0x04080037 0x04090067
+ 0x05010038 0x0502000c 0x0503001b
+ 0x05040034 0x0505001a 0x05060006
+ 0x05080027 0x0509000e 0x050a006f
+ 0x0600002b 0x0602004e 0x06030068
+ 0x06040003 0x0605006d 0x06060009
+ 0x06070001 0x0609000f 0x0708002a
+ 0x0709001d 0x070a0033 >;
+};
+
+/ {
+ model = "TI-NSPIRE Touchpad";
+ compatible = "arm,nspire-tp";
+};
diff --git a/arch/arm/boot/dts/nspire.dtsi b/arch/arm/boot/dts/nspire.dtsi
new file mode 100644
index 0000000..0e780f9
--- /dev/null
+++ b/arch/arm/boot/dts/nspire.dtsi
@@ -0,0 +1,182 @@
+/*
+ * linux/arch/arm/boot/nspire.dtsi
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+/include/ "skeleton.dtsi"
+
+/ {
+ interrupt-parent = <&intc>;
+
+ cpus {
+ cpu at 0 {
+ compatible = "arm,arm926ejs";
+ };
+ };
+
+ bootrom: bootrom at 00000000 {
+ reg = <0x00000000 0x80000>;
+ };
+
+ sram: sram at A4000000 {
+ device = "memory";
+ reg = <0xA4000000 0x20000>;
+ };
+
+ timer_clk: timer_clk {
+ #clock-cells = <0>;
+ compatible = "fixed-clock";
+ clock-frequency = <32768>;
+ };
+
+ base_clk: base_clk {
+ #clock-cells = <0>;
+ compatible = "nspire-clock";
+ reg = <0x900B0000 0x2>;
+ };
+
+ ahb_clk: ahb_clk {
+ #clock-cells = <0>;
+ compatible = "nspire-ahb-divider";
+ reg = <0x900B0000 0x2>;
+ clocks = <&base_clk>;
+ clock-names = "base_clk";
+ };
+
+ apb_pclk: apb_pclk {
+ #clock-cells = <0>;
+ compatible = "fixed-factor-clock";
+ clock-div = <2>;
+ clock-mult = <1>;
+ clocks = <&ahb_clk>;
+ clock-names = "ahb_clk";
+ };
+
+ ahb {
+ compatible = "simple-bus";
+ #address-cells = <1>;
+ #size-cells = <1>;
+ ranges;
+
+ spi: spi at A9000000 {
+ reg = <0xA9000000 0x1000>;
+ };
+
+ usb0: usb at B0000000 {
+ reg = <0xB0000000 0x1000>;
+ interrupts = <8>;
+ };
+
+ usb1: usb at B4000000 {
+ reg = <0xB4000000 0x1000>;
+ interrupts = <9>;
+ status = "disabled";
+ };
+
+ lcd: lcd at C0000000 {
+ compatible = "arm,pl111", "arm,primecell";
+ reg = <0xC0000000 0x1000>;
+ interrupts = <21>;
+
+ clocks = <&apb_pclk>;
+ clock-names = "apb_pclk";
+ };
+
+ adc: adc at C4000000 {
+ reg = <0xC4000000 0x1000>;
+ interrupts = <11>;
+ };
+
+ tdes: crypto at C8010000 {
+ reg = <0xC8010000 0x1000>;
+ };
+
+ sha256: crypto at CC000000 {
+ reg = <0xCC000000 0x1000>;
+ };
+
+ apb at 90000000 {
+ compatible = "simple-bus";
+ #address-cells = <1>;
+ #size-cells = <1>;
+ clock-ranges;
+ ranges;
+
+ gpio: gpio at 90000000 {
+ reg = <0x90000000 0x1000>;
+ interrupts = <7>;
+ };
+
+ fast_timer: timer at 90010000 {
+ reg = <0x90010000 0x1000>;
+ interrupts = <17>;
+ };
+
+ uart: serial at 90020000 {
+ reg = <0x90020000 0x1000>;
+ interrupts = <1>;
+ };
+
+ timer0: timer at 900C0000 {
+ reg = <0x900C0000 0x1000>;
+
+ clocks = <&timer_clk>;
+ clock-names = "timer_clk";
+ };
+
+ timer1: timer at 900D0000 {
+ reg = <0x900D0000 0x1000>;
+ interrupts = <19>;
+
+ clocks = <&timer_clk>;
+ clock-names = "timer_clk";
+ };
+
+ watchdog: watchdog at 90060000 {
+ compatible = "arm,amba-primecell";
+ reg = <0x90060000 0x1000>;
+ interrupts = <3>;
+ };
+
+ rtc: rtc at 90090000 {
+ reg = <0x90090000 0x1000>;
+ interrupts = <4>;
+ };
+
+ misc: misc at 900A0000 {
+ reg = <0x900A0000 0x1000>;
+ };
+
+ pwr: pwr at 900B0000 {
+ reg = <0x900B0000 0x1000>;
+ interrupts = <15>;
+ };
+
+ keypad: input at 900E0000 {
+ compatible = "nspire-keypad";
+ reg = <0x900E0000 0x1000>;
+ interrupts = <16>;
+
+ scan-interval = <1000>;
+ row-delay = <200>;
+
+ clocks = <&apb_pclk>;
+ clock-names = "apb_pclk";
+ };
+
+ contrast: contrast at 900F0000 {
+ reg = <0x900F0000 0x1000>;
+ };
+
+ led: led at 90110000 {
+ reg = <0x90110000 0x1000>;
+ };
+ };
+ };
+};
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-12 4:22 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel, linux
Cc: Daniel Tang, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/clk/Makefile | 1 +
drivers/clk/clk-nspire.c | 141 +++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 142 insertions(+)
create mode 100644 drivers/clk/clk-nspire.c
diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
index 137d3e7..72ebbe1 100644
--- a/drivers/clk/Makefile
+++ b/drivers/clk/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_COMMON_CLK) += clk-composite.o
obj-$(CONFIG_ARCH_BCM2835) += clk-bcm2835.o
obj-$(CONFIG_ARCH_NOMADIK) += clk-nomadik.o
obj-$(CONFIG_ARCH_HIGHBANK) += clk-highbank.o
+obj-$(CONFIG_ARCH_NSPIRE) += clk-nspire.o
obj-$(CONFIG_ARCH_MXS) += mxs/
obj-$(CONFIG_ARCH_SOCFPGA) += socfpga/
obj-$(CONFIG_PLAT_SPEAR) += spear/
diff --git a/drivers/clk/clk-nspire.c b/drivers/clk/clk-nspire.c
new file mode 100644
index 0000000..2a8d14c
--- /dev/null
+++ b/drivers/clk/clk-nspire.c
@@ -0,0 +1,141 @@
+/*
+ * linux/drivers/clk/clk-nspire.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+
+#define MHZ (1000 * 1000)
+
+#define BASE_CPU_SHIFT 1
+#define BASE_CPU_MASK 0x7F
+
+#define CPU_AHB_SHIFT 12
+#define CPU_AHB_MASK 0x07
+
+#define FIXED_BASE_SHIFT 8
+#define FIXED_BASE_MASK 0x01
+
+#define CLASSIC_BASE_SHIFT 16
+#define CLASSIC_BASE_MASK 0x1F
+
+#define CX_BASE_SHIFT 15
+#define CX_BASE_MASK 0x3F
+
+#define CX_UNKNOWN_SHIFT 21
+#define CX_UNKNOWN_MASK 0x03
+
+#define EXTRACT(var, prop) (((var)>>prop##_SHIFT) & prop##_MASK)
+
+struct nspire_clk_info {
+ u32 base_clock;
+ u16 base_cpu_ratio;
+ u16 base_ahb_ratio;
+};
+
+static int nspire_clk_read(struct device_node *node,
+ struct nspire_clk_info *clk)
+{
+ u32 val;
+ int ret;
+ void __iomem *io;
+ const char *type = NULL;
+
+ ret = of_property_read_string(node, "io-type", &type);
+ if (ret)
+ return ret;
+
+ io = of_iomap(node, 0);
+ if (!io)
+ return -ENOMEM;
+ val = readl(io);
+ iounmap(io);
+
+ if (!strcmp(type, "cx")) {
+ if (EXTRACT(val, FIXED_BASE)) {
+ clk->base_clock = 48 * MHZ;
+ } else {
+ clk->base_clock = 6 * EXTRACT(val, CX_BASE) * MHZ;
+ }
+
+ clk->base_cpu_ratio = EXTRACT(val, BASE_CPU) *
+ EXTRACT(val, CX_UNKNOWN);
+ clk->base_ahb_ratio = clk->base_cpu_ratio *
+ (EXTRACT(val, CPU_AHB) + 1);
+ } else if (!strcmp(type, "classic")) {
+ if (EXTRACT(val, FIXED_BASE)) {
+ clk->base_clock = 27 * MHZ;
+ } else {
+ clk->base_clock = (300 -
+ 6 * EXTRACT(val, CLASSIC_BASE)) * MHZ;
+ }
+
+ clk->base_cpu_ratio = EXTRACT(val, BASE_CPU) * 2;
+ clk->base_ahb_ratio = clk->base_cpu_ratio *
+ (EXTRACT(val, CPU_AHB) + 1);
+ } else {
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static void __init nspire_ahbdiv_setup(struct device_node *node)
+{
+ int ret;
+ struct clk *clk;
+ const char *clk_name = node->name;
+ const char *parent_name;
+ struct nspire_clk_info info;
+
+ ret = nspire_clk_read(node, &info);
+ if (WARN_ON(ret))
+ return;
+
+ of_property_read_string(node, "clock-output-names", &clk_name);
+ parent_name = of_clk_get_parent_name(node, 0);
+
+ clk = clk_register_fixed_factor(NULL, clk_name, parent_name, 0,
+ 1, info.base_ahb_ratio);
+ if (!IS_ERR(clk))
+ of_clk_add_provider(node, of_clk_src_simple_get, clk);
+}
+
+static void __init nspire_clk_setup(struct device_node *node)
+{
+ int ret;
+ struct clk *clk;
+ const char *clk_name = node->name;
+ struct nspire_clk_info info;
+
+ ret = nspire_clk_read(node, &info);
+ if (WARN_ON(ret))
+ return;
+
+ of_property_read_string(node, "clock-output-names", &clk_name);
+
+ clk = clk_register_fixed_rate(NULL, clk_name, NULL, CLK_IS_ROOT,
+ info.base_clock);
+ if (!IS_ERR(clk))
+ of_clk_add_provider(node, of_clk_src_simple_get, clk);
+ else
+ return;
+
+ pr_info("TI-NSPIRE Base: %uMHz CPU: %uMHz AHB: %uMHz\n",
+ info.base_clock / MHZ,
+ info.base_clock / info.base_cpu_ratio / MHZ,
+ info.base_clock / info.base_ahb_ratio / MHZ);
+}
+
+CLK_OF_DECLARE(nspire_clk, "nspire-clock", nspire_clk_setup);
+CLK_OF_DECLARE(nspire_ahbdiv, "nspire-ahb-divider", nspire_ahbdiv_setup);
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-12 4:22 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/clk/Makefile | 1 +
drivers/clk/clk-nspire.c | 141 +++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 142 insertions(+)
create mode 100644 drivers/clk/clk-nspire.c
diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
index 137d3e7..72ebbe1 100644
--- a/drivers/clk/Makefile
+++ b/drivers/clk/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_COMMON_CLK) += clk-composite.o
obj-$(CONFIG_ARCH_BCM2835) += clk-bcm2835.o
obj-$(CONFIG_ARCH_NOMADIK) += clk-nomadik.o
obj-$(CONFIG_ARCH_HIGHBANK) += clk-highbank.o
+obj-$(CONFIG_ARCH_NSPIRE) += clk-nspire.o
obj-$(CONFIG_ARCH_MXS) += mxs/
obj-$(CONFIG_ARCH_SOCFPGA) += socfpga/
obj-$(CONFIG_PLAT_SPEAR) += spear/
diff --git a/drivers/clk/clk-nspire.c b/drivers/clk/clk-nspire.c
new file mode 100644
index 0000000..2a8d14c
--- /dev/null
+++ b/drivers/clk/clk-nspire.c
@@ -0,0 +1,141 @@
+/*
+ * linux/drivers/clk/clk-nspire.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+
+#define MHZ (1000 * 1000)
+
+#define BASE_CPU_SHIFT 1
+#define BASE_CPU_MASK 0x7F
+
+#define CPU_AHB_SHIFT 12
+#define CPU_AHB_MASK 0x07
+
+#define FIXED_BASE_SHIFT 8
+#define FIXED_BASE_MASK 0x01
+
+#define CLASSIC_BASE_SHIFT 16
+#define CLASSIC_BASE_MASK 0x1F
+
+#define CX_BASE_SHIFT 15
+#define CX_BASE_MASK 0x3F
+
+#define CX_UNKNOWN_SHIFT 21
+#define CX_UNKNOWN_MASK 0x03
+
+#define EXTRACT(var, prop) (((var)>>prop##_SHIFT) & prop##_MASK)
+
+struct nspire_clk_info {
+ u32 base_clock;
+ u16 base_cpu_ratio;
+ u16 base_ahb_ratio;
+};
+
+static int nspire_clk_read(struct device_node *node,
+ struct nspire_clk_info *clk)
+{
+ u32 val;
+ int ret;
+ void __iomem *io;
+ const char *type = NULL;
+
+ ret = of_property_read_string(node, "io-type", &type);
+ if (ret)
+ return ret;
+
+ io = of_iomap(node, 0);
+ if (!io)
+ return -ENOMEM;
+ val = readl(io);
+ iounmap(io);
+
+ if (!strcmp(type, "cx")) {
+ if (EXTRACT(val, FIXED_BASE)) {
+ clk->base_clock = 48 * MHZ;
+ } else {
+ clk->base_clock = 6 * EXTRACT(val, CX_BASE) * MHZ;
+ }
+
+ clk->base_cpu_ratio = EXTRACT(val, BASE_CPU) *
+ EXTRACT(val, CX_UNKNOWN);
+ clk->base_ahb_ratio = clk->base_cpu_ratio *
+ (EXTRACT(val, CPU_AHB) + 1);
+ } else if (!strcmp(type, "classic")) {
+ if (EXTRACT(val, FIXED_BASE)) {
+ clk->base_clock = 27 * MHZ;
+ } else {
+ clk->base_clock = (300 -
+ 6 * EXTRACT(val, CLASSIC_BASE)) * MHZ;
+ }
+
+ clk->base_cpu_ratio = EXTRACT(val, BASE_CPU) * 2;
+ clk->base_ahb_ratio = clk->base_cpu_ratio *
+ (EXTRACT(val, CPU_AHB) + 1);
+ } else {
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static void __init nspire_ahbdiv_setup(struct device_node *node)
+{
+ int ret;
+ struct clk *clk;
+ const char *clk_name = node->name;
+ const char *parent_name;
+ struct nspire_clk_info info;
+
+ ret = nspire_clk_read(node, &info);
+ if (WARN_ON(ret))
+ return;
+
+ of_property_read_string(node, "clock-output-names", &clk_name);
+ parent_name = of_clk_get_parent_name(node, 0);
+
+ clk = clk_register_fixed_factor(NULL, clk_name, parent_name, 0,
+ 1, info.base_ahb_ratio);
+ if (!IS_ERR(clk))
+ of_clk_add_provider(node, of_clk_src_simple_get, clk);
+}
+
+static void __init nspire_clk_setup(struct device_node *node)
+{
+ int ret;
+ struct clk *clk;
+ const char *clk_name = node->name;
+ struct nspire_clk_info info;
+
+ ret = nspire_clk_read(node, &info);
+ if (WARN_ON(ret))
+ return;
+
+ of_property_read_string(node, "clock-output-names", &clk_name);
+
+ clk = clk_register_fixed_rate(NULL, clk_name, NULL, CLK_IS_ROOT,
+ info.base_clock);
+ if (!IS_ERR(clk))
+ of_clk_add_provider(node, of_clk_src_simple_get, clk);
+ else
+ return;
+
+ pr_info("TI-NSPIRE Base: %uMHz CPU: %uMHz AHB: %uMHz\n",
+ info.base_clock / MHZ,
+ info.base_clock / info.base_cpu_ratio / MHZ,
+ info.base_clock / info.base_ahb_ratio / MHZ);
+}
+
+CLK_OF_DECLARE(nspire_clk, "nspire-clock", nspire_clk_setup);
+CLK_OF_DECLARE(nspire_ahbdiv, "nspire-ahb-divider", nspire_ahbdiv_setup);
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-12 4:22 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel, linux
Cc: Daniel Tang, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/clocksource/Makefile | 1 +
drivers/clocksource/nspire-classic-timer.c | 199 +++++++++++++++++++++++++++++
2 files changed, 200 insertions(+)
create mode 100644 drivers/clocksource/nspire-classic-timer.c
diff --git a/drivers/clocksource/Makefile b/drivers/clocksource/Makefile
index 8d979c7..b9b56cb 100644
--- a/drivers/clocksource/Makefile
+++ b/drivers/clocksource/Makefile
@@ -22,6 +22,7 @@ obj-$(CONFIG_ARCH_PRIMA2) += timer-prima2.o
obj-$(CONFIG_SUN4I_TIMER) += sun4i_timer.o
obj-$(CONFIG_ARCH_TEGRA) += tegra20_timer.o
obj-$(CONFIG_VT8500_TIMER) += vt8500_timer.o
+obj-$(CONFIG_ARCH_NSPIRE) += nspire-classic-timer.o
obj-$(CONFIG_ARCH_BCM) += bcm_kona_timer.o
obj-$(CONFIG_CADENCE_TTC_TIMER) += cadence_ttc_timer.o
obj-$(CONFIG_CLKSRC_EXYNOS_MCT) += exynos_mct.o
diff --git a/drivers/clocksource/nspire-classic-timer.c b/drivers/clocksource/nspire-classic-timer.c
new file mode 100644
index 0000000..4b92b61
--- /dev/null
+++ b/drivers/clocksource/nspire-classic-timer.c
@@ -0,0 +1,199 @@
+/*
+ * linux/drivers/clocksource/nspire-classic-timer.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/clk.h>
+#include <linux/clockchips.h>
+#include <linux/cpumask.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+
+#define DT_COMPAT "nspire-classic-timer"
+
+#define IO_CURRENT_VAL 0x00
+#define IO_DIVIDER 0x04
+#define IO_CONTROL 0x08
+
+#define IO_TIMER1 0x00
+#define IO_TIMER2 0x0C
+
+#define IO_MATCH1 0x18
+#define IO_MATCH2 0x1C
+#define IO_MATCH3 0x20
+#define IO_MATCH4 0x24
+#define IO_MATCH5 0x28
+#define IO_MATCH6 0x2C
+
+#define IO_INTR_STS 0x00
+#define IO_INTR_ACK 0x00
+#define IO_INTR_MSK 0x04
+
+#define CNTL_STOP_TIMER (1<<4)
+#define CNTL_RUN_TIMER (0<<4)
+
+#define CNTL_INC (1<<3)
+#define CNTL_DEC (0<<3)
+
+#define CNTL_TOZERO 0
+#define CNTL_MATCH1 1
+#define CNTL_MATCH2 2
+#define CNTL_MATCH3 3
+#define CNTL_MATCH4 4
+#define CNTL_MATCH5 5
+#define CNTL_MATCH6 6
+#define CNTL_FOREVER 7
+
+struct nspire_timer {
+ void __iomem *base;
+ void __iomem *timer1, *timer2;
+ void __iomem *interrupt_regs;
+
+ int irqnr;
+
+ struct clk *clk;
+ struct clock_event_device clkevt;
+ struct irqaction clkevt_irq;
+
+ char clocksource_name[64];
+ char clockevent_name[64];
+};
+
+static int nspire_timer_set_event(unsigned long delta,
+ struct clock_event_device *dev)
+{
+ unsigned long flags;
+ struct nspire_timer *timer = container_of(dev,
+ struct nspire_timer,
+ clkevt);
+
+ local_irq_save(flags);
+
+ writel(delta, timer->timer1 + IO_CURRENT_VAL);
+ writel(CNTL_RUN_TIMER | CNTL_DEC | CNTL_MATCH1,
+ timer->timer1 + IO_CONTROL);
+
+ local_irq_restore(flags);
+
+ return 0;
+}
+static void nspire_timer_set_mode(enum clock_event_mode mode,
+ struct clock_event_device *evt)
+{
+ evt->mode = mode;
+}
+
+static irqreturn_t nspire_timer_interrupt(int irq, void *dev_id)
+{
+ struct nspire_timer *timer = dev_id;
+
+ writel((1<<0), timer->interrupt_regs + IO_INTR_ACK);
+ writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
+
+ if (timer->clkevt.event_handler)
+ timer->clkevt.event_handler(&timer->clkevt);
+
+ return IRQ_HANDLED;
+}
+
+static int __init nspire_timer_add(struct device_node *node)
+{
+ struct nspire_timer *timer;
+ struct resource res;
+ int ret;
+
+ timer = kzalloc(sizeof(*timer), GFP_ATOMIC);
+ if (!timer)
+ return -ENOMEM;
+
+ timer->base = of_iomap(node, 0);
+ if (!timer->base) {
+ ret = -EINVAL;
+ goto error_free;
+ }
+ timer->timer1 = timer->base + IO_TIMER1;
+ timer->timer2 = timer->base + IO_TIMER2;
+
+ timer->clk = of_clk_get(node, 0);
+ if (IS_ERR(timer->clk)) {
+ ret = PTR_ERR(timer->clk);
+ pr_err("Timer clock not found! (error %d)\n", ret);
+ goto error_unmap;
+ }
+
+ timer->interrupt_regs = of_iomap(node, 1);
+ timer->irqnr = irq_of_parse_and_map(node, 0);
+
+ of_address_to_resource(node, 0, &res);
+ scnprintf(timer->clocksource_name, sizeof(timer->clocksource_name),
+ "%llx.%s_clocksource",
+ (unsigned long long)res.start, node->name);
+
+ scnprintf(timer->clockevent_name, sizeof(timer->clockevent_name),
+ "%llx.%s_clockevent",
+ (unsigned long long)res.start, node->name);
+
+ if (timer->interrupt_regs && timer->irqnr) {
+ timer->clkevt.name = timer->clockevent_name;
+ timer->clkevt.set_next_event = nspire_timer_set_event;
+ timer->clkevt.set_mode = nspire_timer_set_mode;
+ timer->clkevt.rating = 200;
+ timer->clkevt.shift = 32;
+ timer->clkevt.cpumask = cpu_all_mask;
+ timer->clkevt.features =
+ CLOCK_EVT_FEAT_ONESHOT;
+
+ writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
+ writel(0, timer->timer1 + IO_DIVIDER);
+
+ /* Mask out interrupts except the one we're using */
+ writel((1<<0), timer->interrupt_regs + IO_INTR_MSK);
+ writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
+
+ /* Interrupt to occur when timer value matches 0 */
+ writel(0, timer->base + IO_MATCH1);
+
+ timer->clkevt_irq.name = timer->clockevent_name;
+ timer->clkevt_irq.handler = nspire_timer_interrupt;
+ timer->clkevt_irq.dev_id = timer;
+ timer->clkevt_irq.flags =
+ IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL;
+
+ setup_irq(timer->irqnr, &timer->clkevt_irq);
+
+ clockevents_config_and_register(&timer->clkevt,
+ clk_get_rate(timer->clk), 0x0001, 0xfffe);
+ pr_info("Added %s as clockevent\n", timer->clockevent_name);
+ }
+
+ writel(CNTL_RUN_TIMER | CNTL_FOREVER | CNTL_INC,
+ timer->timer2 + IO_CONTROL);
+
+ clocksource_mmio_init(timer->timer2 + IO_CURRENT_VAL,
+ timer->clocksource_name,
+ clk_get_rate(timer->clk),
+ 200, 16,
+ clocksource_mmio_readw_up);
+
+ pr_info("Added %s as clocksource\n", timer->clocksource_name);
+
+ return 0;
+error_unmap:
+ iounmap(timer->base);
+error_free:
+ kfree(timer);
+ return ret;
+}
+
+CLOCKSOURCE_OF_DECLARE(nspire_classic_timer, DT_COMPAT, nspire_timer_add);
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
@ 2013-05-12 4:22 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:22 UTC (permalink / raw)
To: linux-arm-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/clocksource/Makefile | 1 +
drivers/clocksource/nspire-classic-timer.c | 199 +++++++++++++++++++++++++++++
2 files changed, 200 insertions(+)
create mode 100644 drivers/clocksource/nspire-classic-timer.c
diff --git a/drivers/clocksource/Makefile b/drivers/clocksource/Makefile
index 8d979c7..b9b56cb 100644
--- a/drivers/clocksource/Makefile
+++ b/drivers/clocksource/Makefile
@@ -22,6 +22,7 @@ obj-$(CONFIG_ARCH_PRIMA2) += timer-prima2.o
obj-$(CONFIG_SUN4I_TIMER) += sun4i_timer.o
obj-$(CONFIG_ARCH_TEGRA) += tegra20_timer.o
obj-$(CONFIG_VT8500_TIMER) += vt8500_timer.o
+obj-$(CONFIG_ARCH_NSPIRE) += nspire-classic-timer.o
obj-$(CONFIG_ARCH_BCM) += bcm_kona_timer.o
obj-$(CONFIG_CADENCE_TTC_TIMER) += cadence_ttc_timer.o
obj-$(CONFIG_CLKSRC_EXYNOS_MCT) += exynos_mct.o
diff --git a/drivers/clocksource/nspire-classic-timer.c b/drivers/clocksource/nspire-classic-timer.c
new file mode 100644
index 0000000..4b92b61
--- /dev/null
+++ b/drivers/clocksource/nspire-classic-timer.c
@@ -0,0 +1,199 @@
+/*
+ * linux/drivers/clocksource/nspire-classic-timer.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/clk.h>
+#include <linux/clockchips.h>
+#include <linux/cpumask.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+
+#define DT_COMPAT "nspire-classic-timer"
+
+#define IO_CURRENT_VAL 0x00
+#define IO_DIVIDER 0x04
+#define IO_CONTROL 0x08
+
+#define IO_TIMER1 0x00
+#define IO_TIMER2 0x0C
+
+#define IO_MATCH1 0x18
+#define IO_MATCH2 0x1C
+#define IO_MATCH3 0x20
+#define IO_MATCH4 0x24
+#define IO_MATCH5 0x28
+#define IO_MATCH6 0x2C
+
+#define IO_INTR_STS 0x00
+#define IO_INTR_ACK 0x00
+#define IO_INTR_MSK 0x04
+
+#define CNTL_STOP_TIMER (1<<4)
+#define CNTL_RUN_TIMER (0<<4)
+
+#define CNTL_INC (1<<3)
+#define CNTL_DEC (0<<3)
+
+#define CNTL_TOZERO 0
+#define CNTL_MATCH1 1
+#define CNTL_MATCH2 2
+#define CNTL_MATCH3 3
+#define CNTL_MATCH4 4
+#define CNTL_MATCH5 5
+#define CNTL_MATCH6 6
+#define CNTL_FOREVER 7
+
+struct nspire_timer {
+ void __iomem *base;
+ void __iomem *timer1, *timer2;
+ void __iomem *interrupt_regs;
+
+ int irqnr;
+
+ struct clk *clk;
+ struct clock_event_device clkevt;
+ struct irqaction clkevt_irq;
+
+ char clocksource_name[64];
+ char clockevent_name[64];
+};
+
+static int nspire_timer_set_event(unsigned long delta,
+ struct clock_event_device *dev)
+{
+ unsigned long flags;
+ struct nspire_timer *timer = container_of(dev,
+ struct nspire_timer,
+ clkevt);
+
+ local_irq_save(flags);
+
+ writel(delta, timer->timer1 + IO_CURRENT_VAL);
+ writel(CNTL_RUN_TIMER | CNTL_DEC | CNTL_MATCH1,
+ timer->timer1 + IO_CONTROL);
+
+ local_irq_restore(flags);
+
+ return 0;
+}
+static void nspire_timer_set_mode(enum clock_event_mode mode,
+ struct clock_event_device *evt)
+{
+ evt->mode = mode;
+}
+
+static irqreturn_t nspire_timer_interrupt(int irq, void *dev_id)
+{
+ struct nspire_timer *timer = dev_id;
+
+ writel((1<<0), timer->interrupt_regs + IO_INTR_ACK);
+ writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
+
+ if (timer->clkevt.event_handler)
+ timer->clkevt.event_handler(&timer->clkevt);
+
+ return IRQ_HANDLED;
+}
+
+static int __init nspire_timer_add(struct device_node *node)
+{
+ struct nspire_timer *timer;
+ struct resource res;
+ int ret;
+
+ timer = kzalloc(sizeof(*timer), GFP_ATOMIC);
+ if (!timer)
+ return -ENOMEM;
+
+ timer->base = of_iomap(node, 0);
+ if (!timer->base) {
+ ret = -EINVAL;
+ goto error_free;
+ }
+ timer->timer1 = timer->base + IO_TIMER1;
+ timer->timer2 = timer->base + IO_TIMER2;
+
+ timer->clk = of_clk_get(node, 0);
+ if (IS_ERR(timer->clk)) {
+ ret = PTR_ERR(timer->clk);
+ pr_err("Timer clock not found! (error %d)\n", ret);
+ goto error_unmap;
+ }
+
+ timer->interrupt_regs = of_iomap(node, 1);
+ timer->irqnr = irq_of_parse_and_map(node, 0);
+
+ of_address_to_resource(node, 0, &res);
+ scnprintf(timer->clocksource_name, sizeof(timer->clocksource_name),
+ "%llx.%s_clocksource",
+ (unsigned long long)res.start, node->name);
+
+ scnprintf(timer->clockevent_name, sizeof(timer->clockevent_name),
+ "%llx.%s_clockevent",
+ (unsigned long long)res.start, node->name);
+
+ if (timer->interrupt_regs && timer->irqnr) {
+ timer->clkevt.name = timer->clockevent_name;
+ timer->clkevt.set_next_event = nspire_timer_set_event;
+ timer->clkevt.set_mode = nspire_timer_set_mode;
+ timer->clkevt.rating = 200;
+ timer->clkevt.shift = 32;
+ timer->clkevt.cpumask = cpu_all_mask;
+ timer->clkevt.features =
+ CLOCK_EVT_FEAT_ONESHOT;
+
+ writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
+ writel(0, timer->timer1 + IO_DIVIDER);
+
+ /* Mask out interrupts except the one we're using */
+ writel((1<<0), timer->interrupt_regs + IO_INTR_MSK);
+ writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
+
+ /* Interrupt to occur when timer value matches 0 */
+ writel(0, timer->base + IO_MATCH1);
+
+ timer->clkevt_irq.name = timer->clockevent_name;
+ timer->clkevt_irq.handler = nspire_timer_interrupt;
+ timer->clkevt_irq.dev_id = timer;
+ timer->clkevt_irq.flags =
+ IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL;
+
+ setup_irq(timer->irqnr, &timer->clkevt_irq);
+
+ clockevents_config_and_register(&timer->clkevt,
+ clk_get_rate(timer->clk), 0x0001, 0xfffe);
+ pr_info("Added %s as clockevent\n", timer->clockevent_name);
+ }
+
+ writel(CNTL_RUN_TIMER | CNTL_FOREVER | CNTL_INC,
+ timer->timer2 + IO_CONTROL);
+
+ clocksource_mmio_init(timer->timer2 + IO_CURRENT_VAL,
+ timer->clocksource_name,
+ clk_get_rate(timer->clk),
+ 200, 16,
+ clocksource_mmio_readw_up);
+
+ pr_info("Added %s as clocksource\n", timer->clocksource_name);
+
+ return 0;
+error_unmap:
+ iounmap(timer->base);
+error_free:
+ kfree(timer);
+ return ret;
+}
+
+CLOCKSOURCE_OF_DECLARE(nspire_classic_timer, DT_COMPAT, nspire_timer_add);
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 5/6] input: Add TI-Nspire keypad driver
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-12 4:23 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:23 UTC (permalink / raw)
To: linux-arm-kernel, linux
Cc: Daniel Tang, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/input/keyboard/Kconfig | 10 ++
drivers/input/keyboard/Makefile | 1 +
drivers/input/keyboard/nspire-keypad.c | 315 +++++++++++++++++++++++++++++++++
3 files changed, 326 insertions(+)
create mode 100644 drivers/input/keyboard/nspire-keypad.c
diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig
index 62a2c0e..bdbda87 100644
--- a/drivers/input/keyboard/Kconfig
+++ b/drivers/input/keyboard/Kconfig
@@ -418,6 +418,16 @@ config KEYBOARD_NOMADIK
To compile this driver as a module, choose M here: the
module will be called nmk-ske-keypad.
+config KEYBOARD_NSPIRE
+ tristate "TI-NSPIRE builtin keyboard"
+ depends on ARCH_NSPIRE
+ select INPUT_MATRIXKMAP
+ help
+ Say Y here if you want to use the builtin keypad on the TI-NSPIRE.
+
+ To compile this driver as a module, choose M here: the
+ module will be called nspire-keypad.
+
config KEYBOARD_TEGRA
tristate "NVIDIA Tegra internal matrix keyboard controller support"
depends on ARCH_TEGRA && OF
diff --git a/drivers/input/keyboard/Makefile b/drivers/input/keyboard/Makefile
index 0c43e8c..a699b61 100644
--- a/drivers/input/keyboard/Makefile
+++ b/drivers/input/keyboard/Makefile
@@ -36,6 +36,7 @@ obj-$(CONFIG_KEYBOARD_MCS) += mcs_touchkey.o
obj-$(CONFIG_KEYBOARD_MPR121) += mpr121_touchkey.o
obj-$(CONFIG_KEYBOARD_NEWTON) += newtonkbd.o
obj-$(CONFIG_KEYBOARD_NOMADIK) += nomadik-ske-keypad.o
+obj-$(CONFIG_KEYBOARD_NSPIRE) += nspire-keypad.o
obj-$(CONFIG_KEYBOARD_OMAP) += omap-keypad.o
obj-$(CONFIG_KEYBOARD_OMAP4) += omap4-keypad.o
obj-$(CONFIG_KEYBOARD_OPENCORES) += opencores-kbd.o
diff --git a/drivers/input/keyboard/nspire-keypad.c b/drivers/input/keyboard/nspire-keypad.c
new file mode 100644
index 0000000..e863f14
--- /dev/null
+++ b/drivers/input/keyboard/nspire-keypad.c
@@ -0,0 +1,315 @@
+/*
+ * linux/drivers/input/keyboard/nspire-keypad.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/input/matrix_keypad.h>
+#include <linux/platform_device.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/io.h>
+#include <linux/delay.h>
+#include <linux/input.h>
+#include <linux/slab.h>
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/of.h>
+
+#define KEYPAD_SCAN_MODE 0x00
+#define KEYPAD_CNTL 0x04
+#define KEYPAD_INT 0x08
+#define KEYPAD_INTMSK 0x0C
+
+#define KEYPAD_DATA 0x10
+#define KEYPAD_GPIO 0x30
+
+#define KEYPAD_UNKNOWN_INT 0x40
+#define KEYPAD_UNKNOWN_INT_STS 0x44
+
+#define KEYPAD_BITMASK_COLS 11
+#define KEYPAD_BITMASK_ROWS 8
+
+struct nspire_keypad {
+ spinlock_t lock;
+
+ void __iomem *reg_base;
+ int irq;
+ u32 int_mask;
+
+ struct input_dev *input;
+ struct clk *clk;
+
+ struct matrix_keymap_data *keymap;
+ int row_shift;
+
+ /* Maximum delay estimated assuming 33MHz APB */
+ u32 scan_interval; /* In microseconds (~2000us max) */
+ u32 row_delay; /* In microseconds (~500us max) */
+
+ bool active_low;
+};
+
+static inline void nspire_report_state(struct nspire_keypad *keypad,
+ int row, int col, unsigned int state)
+{
+ int code = MATRIX_SCAN_CODE(row, col, keypad->row_shift);
+ unsigned short *keymap = keypad->input->keycode;
+
+ state = keypad->active_low ? !state : !!state;
+ input_report_key(keypad->input, keymap[code], state);
+}
+
+static irqreturn_t nspire_keypad_irq(int irq, void *dev_id)
+{
+ struct nspire_keypad *keypad = dev_id;
+ u32 int_sts;
+ u16 state[8];
+ int row, col;
+
+ int_sts = readl(keypad->reg_base + KEYPAD_INT) & keypad->int_mask;
+
+ if (!int_sts)
+ return IRQ_NONE;
+
+ spin_lock(&keypad->lock);
+
+ memcpy_fromio(state, keypad->reg_base + KEYPAD_DATA, sizeof(state));
+
+ for (row = 0; row < KEYPAD_BITMASK_ROWS; row++) {
+ u16 bits = state[row];
+ for (col = 0; col < KEYPAD_BITMASK_COLS; col++)
+ nspire_report_state(keypad, row, col, bits & (1<<col));
+ }
+ input_sync(keypad->input);
+ writel(0x3, keypad->reg_base + KEYPAD_INT);
+
+ spin_unlock(&keypad->lock);
+
+ return IRQ_HANDLED;
+}
+
+static int nspire_keypad_chip_init(struct nspire_keypad *keypad)
+{
+ unsigned long val = 0, cycles_per_us, delay_cycles, row_delay_cycles;
+
+ cycles_per_us = (clk_get_rate(keypad->clk) / 1000000);
+ if (cycles_per_us == 0)
+ cycles_per_us = 1;
+
+ delay_cycles = cycles_per_us * keypad->scan_interval;
+ WARN_ON(delay_cycles >= (1<<16)); /* Overflow */
+ delay_cycles &= 0xffff;
+
+ row_delay_cycles = cycles_per_us * keypad->row_delay;
+ WARN_ON(row_delay_cycles >= (1<<14)); /* Overflow */
+ row_delay_cycles &= 0x3fff;
+
+
+ val |= (3<<0); /* Set scan mode to 3 (continuous scan) */
+ val |= (row_delay_cycles<<2); /* Delay between scanning each row */
+ val |= (delay_cycles<<16); /* Delay between scans */
+ writel(val, keypad->reg_base + KEYPAD_SCAN_MODE);
+
+ val = (KEYPAD_BITMASK_ROWS & 0xff) | (KEYPAD_BITMASK_COLS & 0xff)<<8;
+ writel(val, keypad->reg_base + KEYPAD_CNTL);
+
+ /* Enable interrupts */
+ keypad->int_mask = (1<<1);
+ writel(keypad->int_mask, keypad->reg_base + 0xc);
+
+ /* Disable GPIO interrupts to prevent hanging on touchpad */
+ /* Possibly used to detect touchpad events */
+ writel(0, keypad->reg_base + KEYPAD_UNKNOWN_INT);
+ /* Acknowledge existing interrupts */
+ writel(~0, keypad->reg_base + KEYPAD_UNKNOWN_INT_STS);
+
+ return 0;
+}
+
+static int nspire_keypad_probe(struct platform_device *pdev)
+{
+ const struct device_node *of_node = pdev->dev.of_node;
+ struct nspire_keypad *keypad;
+ struct input_dev *input;
+ struct resource *res;
+ struct clk *clk;
+ int irq, error;
+
+ irq = platform_get_irq(pdev, 0);
+ if (irq < 0) {
+ dev_err(&pdev->dev, "failed to get keypad irq\n");
+ return -EINVAL;
+ }
+
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "missing platform resources\n");
+ return -EINVAL;
+ }
+
+ clk = clk_get(&pdev->dev, NULL);
+ if (IS_ERR(clk)) {
+ dev_err(&pdev->dev, "unable to get clock\n");
+ return -EINVAL;
+ }
+
+ clk_prepare(clk);
+ error = clk_enable(clk);
+ if (error)
+ goto err_put_clk;
+
+ keypad = kzalloc(sizeof(struct nspire_keypad), GFP_KERNEL);
+ input = input_allocate_device();
+ if (!keypad || !input) {
+ dev_err(&pdev->dev, "failed to allocate keypad memory\n");
+ error = -ENOMEM;
+ goto err_free_mem;
+ }
+
+ error = of_property_read_u32(of_node, "scan-interval",
+ &keypad->scan_interval);
+ if (error) {
+ dev_err(&pdev->dev, "failed to get scan-interval\n");
+ goto err_free_mem;
+ }
+
+ error = of_property_read_u32(of_node, "row-delay",
+ &keypad->row_delay);
+ if (error) {
+ dev_err(&pdev->dev, "failed to get row-delay\n");
+ goto err_free_mem;
+ }
+
+ keypad->active_low = of_property_read_bool(of_node,
+ "active-low");
+
+ keypad->keymap = NULL;
+
+ keypad->row_shift = get_count_order(KEYPAD_BITMASK_COLS);
+ keypad->irq = irq;
+ keypad->clk = clk;
+ keypad->input = input;
+ spin_lock_init(&keypad->lock);
+
+ if (!request_mem_region(res->start, resource_size(res), pdev->name)) {
+ dev_err(&pdev->dev, "failed to request I/O memory\n");
+ error = -EBUSY;
+ goto err_free_mem;
+ }
+
+ keypad->reg_base = ioremap(res->start, resource_size(res));
+ if (!keypad->reg_base) {
+ dev_err(&pdev->dev, "failed to remap I/O memory\n");
+ error = -ENXIO;
+ goto err_free_mem_region;
+ }
+
+ input->id.bustype = BUS_HOST;
+ input->name = "nspire-keypad";
+ input->dev.parent = &pdev->dev;
+
+ set_bit(EV_KEY, input->evbit);
+ set_bit(EV_REP, input->evbit);
+
+ error = matrix_keypad_build_keymap(keypad->keymap, "keymap",
+ KEYPAD_BITMASK_ROWS, KEYPAD_BITMASK_COLS, NULL, input);
+ if (error) {
+ dev_err(&pdev->dev, "building keymap failed\n");
+ goto err_iounmap;
+ }
+
+ error = nspire_keypad_chip_init(keypad);
+ if (error) {
+ dev_err(&pdev->dev, "unable to init keypad hardware\n");
+ goto err_iounmap;
+ }
+
+ error = request_irq(keypad->irq, nspire_keypad_irq, 0,
+ "nspire_keypad", keypad);
+ if (error) {
+ dev_err(&pdev->dev, "allocate irq %d failed\n", keypad->irq);
+ goto err_iounmap;
+ }
+
+ error = input_register_device(input);
+ if (error) {
+ dev_err(&pdev->dev,
+ "unable to register input device: %d\n", error);
+ goto err_free_irq;
+ }
+
+ platform_set_drvdata(pdev, keypad);
+
+ dev_info(&pdev->dev, "TI-NSPIRE keypad at %#08x ("
+ "scan_interval=%uus,row_delay=%uus"
+ "%s)\n", res->start,
+ keypad->row_delay,
+ keypad->scan_interval,
+ keypad->active_low ? ",active_low" : "");
+
+ return 0;
+
+err_free_irq:
+ free_irq(keypad->irq, keypad);
+err_iounmap:
+ iounmap(keypad->reg_base);
+err_free_mem_region:
+ release_mem_region(res->start, resource_size(res));
+err_free_mem:
+ input_free_device(input);
+ kfree(keypad);
+
+ clk_disable(clk);
+err_put_clk:
+ clk_unprepare(clk);
+ clk_put(clk);
+ return error;
+}
+
+static int nspire_keypad_remove(struct platform_device *pdev)
+{
+ struct nspire_keypad *keypad = platform_get_drvdata(pdev);
+ struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+ free_irq(keypad->irq, keypad);
+
+ input_unregister_device(keypad->input);
+
+ iounmap(keypad->reg_base);
+ release_mem_region(res->start, resource_size(res));
+ kfree(keypad);
+
+ clk_disable(keypad->clk);
+ clk_unprepare(keypad->clk);
+ clk_put(keypad->clk);
+
+ return 0;
+}
+#ifdef CONFIG_OF
+static const struct of_device_id nspire_keypad_dt_match[] = {
+ { .compatible = "nspire-keypad" },
+ { },
+};
+MODULE_DEVICE_TABLE(of, nspire_keypad_dt_match);
+#endif
+
+static struct platform_driver nspire_keypad_driver = {
+ .driver = {
+ .name = "nspire-keypad",
+ .owner = THIS_MODULE,
+ .of_match_table = of_match_ptr(nspire_keypad_dt_match),
+ },
+ .remove = nspire_keypad_remove,
+ .probe = nspire_keypad_probe,
+};
+
+module_platform_driver(nspire_keypad_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("TI-NSPIRE Keypad Driver");
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 5/6] input: Add TI-Nspire keypad driver
@ 2013-05-12 4:23 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:23 UTC (permalink / raw)
To: linux-arm-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/input/keyboard/Kconfig | 10 ++
drivers/input/keyboard/Makefile | 1 +
drivers/input/keyboard/nspire-keypad.c | 315 +++++++++++++++++++++++++++++++++
3 files changed, 326 insertions(+)
create mode 100644 drivers/input/keyboard/nspire-keypad.c
diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig
index 62a2c0e..bdbda87 100644
--- a/drivers/input/keyboard/Kconfig
+++ b/drivers/input/keyboard/Kconfig
@@ -418,6 +418,16 @@ config KEYBOARD_NOMADIK
To compile this driver as a module, choose M here: the
module will be called nmk-ske-keypad.
+config KEYBOARD_NSPIRE
+ tristate "TI-NSPIRE builtin keyboard"
+ depends on ARCH_NSPIRE
+ select INPUT_MATRIXKMAP
+ help
+ Say Y here if you want to use the builtin keypad on the TI-NSPIRE.
+
+ To compile this driver as a module, choose M here: the
+ module will be called nspire-keypad.
+
config KEYBOARD_TEGRA
tristate "NVIDIA Tegra internal matrix keyboard controller support"
depends on ARCH_TEGRA && OF
diff --git a/drivers/input/keyboard/Makefile b/drivers/input/keyboard/Makefile
index 0c43e8c..a699b61 100644
--- a/drivers/input/keyboard/Makefile
+++ b/drivers/input/keyboard/Makefile
@@ -36,6 +36,7 @@ obj-$(CONFIG_KEYBOARD_MCS) += mcs_touchkey.o
obj-$(CONFIG_KEYBOARD_MPR121) += mpr121_touchkey.o
obj-$(CONFIG_KEYBOARD_NEWTON) += newtonkbd.o
obj-$(CONFIG_KEYBOARD_NOMADIK) += nomadik-ske-keypad.o
+obj-$(CONFIG_KEYBOARD_NSPIRE) += nspire-keypad.o
obj-$(CONFIG_KEYBOARD_OMAP) += omap-keypad.o
obj-$(CONFIG_KEYBOARD_OMAP4) += omap4-keypad.o
obj-$(CONFIG_KEYBOARD_OPENCORES) += opencores-kbd.o
diff --git a/drivers/input/keyboard/nspire-keypad.c b/drivers/input/keyboard/nspire-keypad.c
new file mode 100644
index 0000000..e863f14
--- /dev/null
+++ b/drivers/input/keyboard/nspire-keypad.c
@@ -0,0 +1,315 @@
+/*
+ * linux/drivers/input/keyboard/nspire-keypad.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/input/matrix_keypad.h>
+#include <linux/platform_device.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/io.h>
+#include <linux/delay.h>
+#include <linux/input.h>
+#include <linux/slab.h>
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/of.h>
+
+#define KEYPAD_SCAN_MODE 0x00
+#define KEYPAD_CNTL 0x04
+#define KEYPAD_INT 0x08
+#define KEYPAD_INTMSK 0x0C
+
+#define KEYPAD_DATA 0x10
+#define KEYPAD_GPIO 0x30
+
+#define KEYPAD_UNKNOWN_INT 0x40
+#define KEYPAD_UNKNOWN_INT_STS 0x44
+
+#define KEYPAD_BITMASK_COLS 11
+#define KEYPAD_BITMASK_ROWS 8
+
+struct nspire_keypad {
+ spinlock_t lock;
+
+ void __iomem *reg_base;
+ int irq;
+ u32 int_mask;
+
+ struct input_dev *input;
+ struct clk *clk;
+
+ struct matrix_keymap_data *keymap;
+ int row_shift;
+
+ /* Maximum delay estimated assuming 33MHz APB */
+ u32 scan_interval; /* In microseconds (~2000us max) */
+ u32 row_delay; /* In microseconds (~500us max) */
+
+ bool active_low;
+};
+
+static inline void nspire_report_state(struct nspire_keypad *keypad,
+ int row, int col, unsigned int state)
+{
+ int code = MATRIX_SCAN_CODE(row, col, keypad->row_shift);
+ unsigned short *keymap = keypad->input->keycode;
+
+ state = keypad->active_low ? !state : !!state;
+ input_report_key(keypad->input, keymap[code], state);
+}
+
+static irqreturn_t nspire_keypad_irq(int irq, void *dev_id)
+{
+ struct nspire_keypad *keypad = dev_id;
+ u32 int_sts;
+ u16 state[8];
+ int row, col;
+
+ int_sts = readl(keypad->reg_base + KEYPAD_INT) & keypad->int_mask;
+
+ if (!int_sts)
+ return IRQ_NONE;
+
+ spin_lock(&keypad->lock);
+
+ memcpy_fromio(state, keypad->reg_base + KEYPAD_DATA, sizeof(state));
+
+ for (row = 0; row < KEYPAD_BITMASK_ROWS; row++) {
+ u16 bits = state[row];
+ for (col = 0; col < KEYPAD_BITMASK_COLS; col++)
+ nspire_report_state(keypad, row, col, bits & (1<<col));
+ }
+ input_sync(keypad->input);
+ writel(0x3, keypad->reg_base + KEYPAD_INT);
+
+ spin_unlock(&keypad->lock);
+
+ return IRQ_HANDLED;
+}
+
+static int nspire_keypad_chip_init(struct nspire_keypad *keypad)
+{
+ unsigned long val = 0, cycles_per_us, delay_cycles, row_delay_cycles;
+
+ cycles_per_us = (clk_get_rate(keypad->clk) / 1000000);
+ if (cycles_per_us == 0)
+ cycles_per_us = 1;
+
+ delay_cycles = cycles_per_us * keypad->scan_interval;
+ WARN_ON(delay_cycles >= (1<<16)); /* Overflow */
+ delay_cycles &= 0xffff;
+
+ row_delay_cycles = cycles_per_us * keypad->row_delay;
+ WARN_ON(row_delay_cycles >= (1<<14)); /* Overflow */
+ row_delay_cycles &= 0x3fff;
+
+
+ val |= (3<<0); /* Set scan mode to 3 (continuous scan) */
+ val |= (row_delay_cycles<<2); /* Delay between scanning each row */
+ val |= (delay_cycles<<16); /* Delay between scans */
+ writel(val, keypad->reg_base + KEYPAD_SCAN_MODE);
+
+ val = (KEYPAD_BITMASK_ROWS & 0xff) | (KEYPAD_BITMASK_COLS & 0xff)<<8;
+ writel(val, keypad->reg_base + KEYPAD_CNTL);
+
+ /* Enable interrupts */
+ keypad->int_mask = (1<<1);
+ writel(keypad->int_mask, keypad->reg_base + 0xc);
+
+ /* Disable GPIO interrupts to prevent hanging on touchpad */
+ /* Possibly used to detect touchpad events */
+ writel(0, keypad->reg_base + KEYPAD_UNKNOWN_INT);
+ /* Acknowledge existing interrupts */
+ writel(~0, keypad->reg_base + KEYPAD_UNKNOWN_INT_STS);
+
+ return 0;
+}
+
+static int nspire_keypad_probe(struct platform_device *pdev)
+{
+ const struct device_node *of_node = pdev->dev.of_node;
+ struct nspire_keypad *keypad;
+ struct input_dev *input;
+ struct resource *res;
+ struct clk *clk;
+ int irq, error;
+
+ irq = platform_get_irq(pdev, 0);
+ if (irq < 0) {
+ dev_err(&pdev->dev, "failed to get keypad irq\n");
+ return -EINVAL;
+ }
+
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "missing platform resources\n");
+ return -EINVAL;
+ }
+
+ clk = clk_get(&pdev->dev, NULL);
+ if (IS_ERR(clk)) {
+ dev_err(&pdev->dev, "unable to get clock\n");
+ return -EINVAL;
+ }
+
+ clk_prepare(clk);
+ error = clk_enable(clk);
+ if (error)
+ goto err_put_clk;
+
+ keypad = kzalloc(sizeof(struct nspire_keypad), GFP_KERNEL);
+ input = input_allocate_device();
+ if (!keypad || !input) {
+ dev_err(&pdev->dev, "failed to allocate keypad memory\n");
+ error = -ENOMEM;
+ goto err_free_mem;
+ }
+
+ error = of_property_read_u32(of_node, "scan-interval",
+ &keypad->scan_interval);
+ if (error) {
+ dev_err(&pdev->dev, "failed to get scan-interval\n");
+ goto err_free_mem;
+ }
+
+ error = of_property_read_u32(of_node, "row-delay",
+ &keypad->row_delay);
+ if (error) {
+ dev_err(&pdev->dev, "failed to get row-delay\n");
+ goto err_free_mem;
+ }
+
+ keypad->active_low = of_property_read_bool(of_node,
+ "active-low");
+
+ keypad->keymap = NULL;
+
+ keypad->row_shift = get_count_order(KEYPAD_BITMASK_COLS);
+ keypad->irq = irq;
+ keypad->clk = clk;
+ keypad->input = input;
+ spin_lock_init(&keypad->lock);
+
+ if (!request_mem_region(res->start, resource_size(res), pdev->name)) {
+ dev_err(&pdev->dev, "failed to request I/O memory\n");
+ error = -EBUSY;
+ goto err_free_mem;
+ }
+
+ keypad->reg_base = ioremap(res->start, resource_size(res));
+ if (!keypad->reg_base) {
+ dev_err(&pdev->dev, "failed to remap I/O memory\n");
+ error = -ENXIO;
+ goto err_free_mem_region;
+ }
+
+ input->id.bustype = BUS_HOST;
+ input->name = "nspire-keypad";
+ input->dev.parent = &pdev->dev;
+
+ set_bit(EV_KEY, input->evbit);
+ set_bit(EV_REP, input->evbit);
+
+ error = matrix_keypad_build_keymap(keypad->keymap, "keymap",
+ KEYPAD_BITMASK_ROWS, KEYPAD_BITMASK_COLS, NULL, input);
+ if (error) {
+ dev_err(&pdev->dev, "building keymap failed\n");
+ goto err_iounmap;
+ }
+
+ error = nspire_keypad_chip_init(keypad);
+ if (error) {
+ dev_err(&pdev->dev, "unable to init keypad hardware\n");
+ goto err_iounmap;
+ }
+
+ error = request_irq(keypad->irq, nspire_keypad_irq, 0,
+ "nspire_keypad", keypad);
+ if (error) {
+ dev_err(&pdev->dev, "allocate irq %d failed\n", keypad->irq);
+ goto err_iounmap;
+ }
+
+ error = input_register_device(input);
+ if (error) {
+ dev_err(&pdev->dev,
+ "unable to register input device: %d\n", error);
+ goto err_free_irq;
+ }
+
+ platform_set_drvdata(pdev, keypad);
+
+ dev_info(&pdev->dev, "TI-NSPIRE keypad at %#08x ("
+ "scan_interval=%uus,row_delay=%uus"
+ "%s)\n", res->start,
+ keypad->row_delay,
+ keypad->scan_interval,
+ keypad->active_low ? ",active_low" : "");
+
+ return 0;
+
+err_free_irq:
+ free_irq(keypad->irq, keypad);
+err_iounmap:
+ iounmap(keypad->reg_base);
+err_free_mem_region:
+ release_mem_region(res->start, resource_size(res));
+err_free_mem:
+ input_free_device(input);
+ kfree(keypad);
+
+ clk_disable(clk);
+err_put_clk:
+ clk_unprepare(clk);
+ clk_put(clk);
+ return error;
+}
+
+static int nspire_keypad_remove(struct platform_device *pdev)
+{
+ struct nspire_keypad *keypad = platform_get_drvdata(pdev);
+ struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+ free_irq(keypad->irq, keypad);
+
+ input_unregister_device(keypad->input);
+
+ iounmap(keypad->reg_base);
+ release_mem_region(res->start, resource_size(res));
+ kfree(keypad);
+
+ clk_disable(keypad->clk);
+ clk_unprepare(keypad->clk);
+ clk_put(keypad->clk);
+
+ return 0;
+}
+#ifdef CONFIG_OF
+static const struct of_device_id nspire_keypad_dt_match[] = {
+ { .compatible = "nspire-keypad" },
+ { },
+};
+MODULE_DEVICE_TABLE(of, nspire_keypad_dt_match);
+#endif
+
+static struct platform_driver nspire_keypad_driver = {
+ .driver = {
+ .name = "nspire-keypad",
+ .owner = THIS_MODULE,
+ .of_match_table = of_match_ptr(nspire_keypad_dt_match),
+ },
+ .remove = nspire_keypad_remove,
+ .probe = nspire_keypad_probe,
+};
+
+module_platform_driver(nspire_keypad_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("TI-NSPIRE Keypad Driver");
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 6/6] irqchip: Add TI-Nspire irqchip
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-12 4:23 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:23 UTC (permalink / raw)
To: linux-arm-kernel, linux
Cc: Daniel Tang, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/irqchip/Makefile | 1 +
drivers/irqchip/irq-nspire-classic.c | 177 +++++++++++++++++++++++++++++++++++
2 files changed, 178 insertions(+)
create mode 100644 drivers/irqchip/irq-nspire-classic.c
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index cda4cb5..056ad7d 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -15,4 +15,5 @@ obj-$(CONFIG_SIRF_IRQ) += irq-sirfsoc.o
obj-$(CONFIG_RENESAS_INTC_IRQPIN) += irq-renesas-intc-irqpin.o
obj-$(CONFIG_RENESAS_IRQC) += irq-renesas-irqc.o
obj-$(CONFIG_VERSATILE_FPGA_IRQ) += irq-versatile-fpga.o
+obj-$(CONFIG_ARCH_NSPIRE) += irq-nspire-classic.o
obj-$(CONFIG_ARCH_VT8500) += irq-vt8500.o
diff --git a/drivers/irqchip/irq-nspire-classic.c b/drivers/irqchip/irq-nspire-classic.c
new file mode 100644
index 0000000..9e6413a
--- /dev/null
+++ b/drivers/irqchip/irq-nspire-classic.c
@@ -0,0 +1,177 @@
+/*
+ * linux/drivers/irqchip/irq-nspire-classic.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+
+#include <asm/mach/irq.h>
+#include <asm/exception.h>
+
+#include "irqchip.h"
+
+#define IO_STATUS 0x000
+#define IO_RAW_STATUS 0x004
+#define IO_ENABLE 0x008
+#define IO_DISABLE 0x00C
+#define IO_CURRENT 0x020
+#define IO_RESET 0x028
+#define IO_MAX_PRIOTY 0x02C
+
+#define IO_IRQ_BASE 0x000
+#define IO_FIQ_BASE 0x100
+
+#define IO_INVERT_SEL 0x200
+#define IO_STICKY_SEL 0x204
+#define IO_PRIORITY_SEL 0x300
+
+#define MAX_INTRS 32
+#define FIQ_START MAX_INTRS
+
+
+static void __iomem *irq_io_base;
+static struct irq_domain *nspire_irq_domain;
+
+static void nspire_irq_ack(struct irq_data *irqd)
+{
+ void __iomem *base = irq_io_base;
+
+ if (irqd->hwirq < FIQ_START)
+ base += IO_IRQ_BASE;
+ else
+ base += IO_FIQ_BASE;
+
+ readl(base + IO_RESET);
+}
+
+static void nspire_irq_unmask(struct irq_data *irqd)
+{
+ void __iomem *base = irq_io_base;
+ int irqnr = irqd->hwirq;
+
+ if (irqnr < FIQ_START) {
+ base += IO_IRQ_BASE;
+ } else {
+ irqnr -= MAX_INTRS;
+ base += IO_FIQ_BASE;
+ }
+
+ writel((1<<irqnr), base + IO_ENABLE);
+}
+
+static void nspire_irq_mask(struct irq_data *irqd)
+{
+ void __iomem *base = irq_io_base;
+ int irqnr = irqd->hwirq;
+
+ if (irqnr < FIQ_START) {
+ base += IO_IRQ_BASE;
+ } else {
+ irqnr -= FIQ_START;
+ base += IO_FIQ_BASE;
+ }
+
+ writel((1<<irqnr), base + IO_DISABLE);
+}
+
+static struct irq_chip nspire_irq_chip = {
+ .name = "nspire_irq",
+ .irq_ack = nspire_irq_ack,
+ .irq_mask = nspire_irq_mask,
+ .irq_unmask = nspire_irq_unmask,
+};
+
+
+static int nspire_irq_map(struct irq_domain *dom, unsigned int virq,
+ irq_hw_number_t hw)
+{
+ irq_set_chip_and_handler(virq, &nspire_irq_chip, handle_level_irq);
+ set_irq_flags(virq, IRQF_VALID | IRQF_PROBE);
+
+ return 0;
+}
+
+static struct irq_domain_ops nspire_irq_ops = {
+ .map = nspire_irq_map,
+ .xlate = irq_domain_xlate_onecell,
+};
+
+static void init_base(void __iomem *base)
+{
+ /* Disable all interrupts */
+ writel(~0, base + IO_DISABLE);
+
+ /* Accept interrupts of all priorities */
+ writel(0xF, base + IO_MAX_PRIOTY);
+
+ /* Reset existing interrupts */
+ readl(base + IO_RESET);
+}
+
+static int process_base(void __iomem *base, struct pt_regs *regs)
+{
+ int irqnr;
+
+
+ if (!readl(base + IO_STATUS))
+ return 0;
+
+ irqnr = readl(base + IO_CURRENT);
+ irqnr = irq_find_mapping(nspire_irq_domain, irqnr);
+ handle_IRQ(irqnr, regs);
+
+ return 1;
+}
+
+asmlinkage void __exception_irq_entry nspire_handle_irq(struct pt_regs *regs)
+{
+ while (process_base(irq_io_base + IO_FIQ_BASE, regs))
+ ;
+ while (process_base(irq_io_base + IO_IRQ_BASE, regs))
+ ;
+}
+
+static int __init nspire_of_init(struct device_node *node,
+ struct device_node *parent)
+{
+ if (WARN_ON(irq_io_base))
+ return -EBUSY;
+
+ irq_io_base = of_iomap(node, 0);
+ BUG_ON(!irq_io_base);
+
+ /* Do not invert interrupt status bits */
+ writel(~0, irq_io_base + IO_INVERT_SEL);
+
+ /* Disable sticky interrupts */
+ writel(0, irq_io_base + IO_STICKY_SEL);
+
+ /* We don't use IRQ priorities. Set each IRQ to highest priority. */
+ memset_io(irq_io_base + IO_PRIORITY_SEL, 0, MAX_INTRS * sizeof(u32));
+
+ /* Init IRQ and FIQ */
+ init_base(irq_io_base + IO_IRQ_BASE);
+ init_base(irq_io_base + IO_FIQ_BASE);
+
+ nspire_irq_domain = irq_domain_add_linear(node, MAX_INTRS,
+ &nspire_irq_ops, NULL);
+
+ BUG_ON(!nspire_irq_domain);
+
+ set_handle_irq(nspire_handle_irq);
+
+ pr_info("TI-NSPIRE classic IRQ controller\n");
+ return 0;
+}
+
+IRQCHIP_DECLARE(nspire_classic_irq, "nspire-classic-intc", nspire_of_init);
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* [RFC PATCHv3 6/6] irqchip: Add TI-Nspire irqchip
@ 2013-05-12 4:23 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 4:23 UTC (permalink / raw)
To: linux-arm-kernel
Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
---
drivers/irqchip/Makefile | 1 +
drivers/irqchip/irq-nspire-classic.c | 177 +++++++++++++++++++++++++++++++++++
2 files changed, 178 insertions(+)
create mode 100644 drivers/irqchip/irq-nspire-classic.c
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index cda4cb5..056ad7d 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -15,4 +15,5 @@ obj-$(CONFIG_SIRF_IRQ) += irq-sirfsoc.o
obj-$(CONFIG_RENESAS_INTC_IRQPIN) += irq-renesas-intc-irqpin.o
obj-$(CONFIG_RENESAS_IRQC) += irq-renesas-irqc.o
obj-$(CONFIG_VERSATILE_FPGA_IRQ) += irq-versatile-fpga.o
+obj-$(CONFIG_ARCH_NSPIRE) += irq-nspire-classic.o
obj-$(CONFIG_ARCH_VT8500) += irq-vt8500.o
diff --git a/drivers/irqchip/irq-nspire-classic.c b/drivers/irqchip/irq-nspire-classic.c
new file mode 100644
index 0000000..9e6413a
--- /dev/null
+++ b/drivers/irqchip/irq-nspire-classic.c
@@ -0,0 +1,177 @@
+/*
+ * linux/drivers/irqchip/irq-nspire-classic.c
+ *
+ * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+
+#include <asm/mach/irq.h>
+#include <asm/exception.h>
+
+#include "irqchip.h"
+
+#define IO_STATUS 0x000
+#define IO_RAW_STATUS 0x004
+#define IO_ENABLE 0x008
+#define IO_DISABLE 0x00C
+#define IO_CURRENT 0x020
+#define IO_RESET 0x028
+#define IO_MAX_PRIOTY 0x02C
+
+#define IO_IRQ_BASE 0x000
+#define IO_FIQ_BASE 0x100
+
+#define IO_INVERT_SEL 0x200
+#define IO_STICKY_SEL 0x204
+#define IO_PRIORITY_SEL 0x300
+
+#define MAX_INTRS 32
+#define FIQ_START MAX_INTRS
+
+
+static void __iomem *irq_io_base;
+static struct irq_domain *nspire_irq_domain;
+
+static void nspire_irq_ack(struct irq_data *irqd)
+{
+ void __iomem *base = irq_io_base;
+
+ if (irqd->hwirq < FIQ_START)
+ base += IO_IRQ_BASE;
+ else
+ base += IO_FIQ_BASE;
+
+ readl(base + IO_RESET);
+}
+
+static void nspire_irq_unmask(struct irq_data *irqd)
+{
+ void __iomem *base = irq_io_base;
+ int irqnr = irqd->hwirq;
+
+ if (irqnr < FIQ_START) {
+ base += IO_IRQ_BASE;
+ } else {
+ irqnr -= MAX_INTRS;
+ base += IO_FIQ_BASE;
+ }
+
+ writel((1<<irqnr), base + IO_ENABLE);
+}
+
+static void nspire_irq_mask(struct irq_data *irqd)
+{
+ void __iomem *base = irq_io_base;
+ int irqnr = irqd->hwirq;
+
+ if (irqnr < FIQ_START) {
+ base += IO_IRQ_BASE;
+ } else {
+ irqnr -= FIQ_START;
+ base += IO_FIQ_BASE;
+ }
+
+ writel((1<<irqnr), base + IO_DISABLE);
+}
+
+static struct irq_chip nspire_irq_chip = {
+ .name = "nspire_irq",
+ .irq_ack = nspire_irq_ack,
+ .irq_mask = nspire_irq_mask,
+ .irq_unmask = nspire_irq_unmask,
+};
+
+
+static int nspire_irq_map(struct irq_domain *dom, unsigned int virq,
+ irq_hw_number_t hw)
+{
+ irq_set_chip_and_handler(virq, &nspire_irq_chip, handle_level_irq);
+ set_irq_flags(virq, IRQF_VALID | IRQF_PROBE);
+
+ return 0;
+}
+
+static struct irq_domain_ops nspire_irq_ops = {
+ .map = nspire_irq_map,
+ .xlate = irq_domain_xlate_onecell,
+};
+
+static void init_base(void __iomem *base)
+{
+ /* Disable all interrupts */
+ writel(~0, base + IO_DISABLE);
+
+ /* Accept interrupts of all priorities */
+ writel(0xF, base + IO_MAX_PRIOTY);
+
+ /* Reset existing interrupts */
+ readl(base + IO_RESET);
+}
+
+static int process_base(void __iomem *base, struct pt_regs *regs)
+{
+ int irqnr;
+
+
+ if (!readl(base + IO_STATUS))
+ return 0;
+
+ irqnr = readl(base + IO_CURRENT);
+ irqnr = irq_find_mapping(nspire_irq_domain, irqnr);
+ handle_IRQ(irqnr, regs);
+
+ return 1;
+}
+
+asmlinkage void __exception_irq_entry nspire_handle_irq(struct pt_regs *regs)
+{
+ while (process_base(irq_io_base + IO_FIQ_BASE, regs))
+ ;
+ while (process_base(irq_io_base + IO_IRQ_BASE, regs))
+ ;
+}
+
+static int __init nspire_of_init(struct device_node *node,
+ struct device_node *parent)
+{
+ if (WARN_ON(irq_io_base))
+ return -EBUSY;
+
+ irq_io_base = of_iomap(node, 0);
+ BUG_ON(!irq_io_base);
+
+ /* Do not invert interrupt status bits */
+ writel(~0, irq_io_base + IO_INVERT_SEL);
+
+ /* Disable sticky interrupts */
+ writel(0, irq_io_base + IO_STICKY_SEL);
+
+ /* We don't use IRQ priorities. Set each IRQ to highest priority. */
+ memset_io(irq_io_base + IO_PRIORITY_SEL, 0, MAX_INTRS * sizeof(u32));
+
+ /* Init IRQ and FIQ */
+ init_base(irq_io_base + IO_IRQ_BASE);
+ init_base(irq_io_base + IO_FIQ_BASE);
+
+ nspire_irq_domain = irq_domain_add_linear(node, MAX_INTRS,
+ &nspire_irq_ops, NULL);
+
+ BUG_ON(!nspire_irq_domain);
+
+ set_handle_irq(nspire_handle_irq);
+
+ pr_info("TI-NSPIRE classic IRQ controller\n");
+ return 0;
+}
+
+IRQCHIP_DECLARE(nspire_classic_irq, "nspire-classic-intc", nspire_of_init);
--
1.8.1.3
^ permalink raw reply related [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 1/6] arm: Initial TI-Nspire support
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-12 9:06 ` Russell King - ARM Linux
-1 siblings, 0 replies; 44+ messages in thread
From: Russell King - ARM Linux @ 2013-05-12 9:06 UTC (permalink / raw)
To: Daniel Tang
Cc: linux-arm-kernel, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On Sun, May 12, 2013 at 02:22:56PM +1000, Daniel Tang wrote:
> diff --git a/arch/arm/mach-nspire/clcd.c b/arch/arm/mach-nspire/clcd.c
> new file mode 100644
> index 0000000..a4b9f06
> --- /dev/null
> +++ b/arch/arm/mach-nspire/clcd.c
> @@ -0,0 +1,117 @@
> +/*
> + * linux/arch/arm/mach-nspire/clcd.c
> + *
> + * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2, as
> + * published by the Free Software Foundation.
> + *
> + */
> +
> +#include <linux/init.h>
> +#include <linux/of.h>
> +#include <linux/amba/bus.h>
> +#include <linux/amba/clcd.h>
> +#include <linux/dma-mapping.h>
> +
> +static struct clcd_panel nspire_cx_lcd_panel = {
> + .mode = {
> + .name = "Color LCD",
> + .refresh = 60,
> + .xres = 320,
> + .yres = 240,
> + .sync = 0,
> + .vmode = FB_VMODE_NONINTERLACED,
> + .pixclock = 1,
> + .hsync_len = 6,
> + .vsync_len = 1,
> + .right_margin = 50,
> + .left_margin = 38,
> + .lower_margin = 3,
> + .upper_margin = 17,
> + },
> + .width = 65, /* ~6.50 cm */
> + .height = 49, /* ~4.87 cm */
> + .tim2 = TIM2_IPC,
> + .cntl = CNTL_LCDTFT | CNTL_LCDVCOMP(1),
> + .bpp = 16,
> +};
> +
> +static struct clcd_panel nspire_classic_lcd_panel = {
> + .mode = {
> + .name = "Grayscale LCD",
> + .refresh = 60,
> + .xres = 320,
> + .yres = 240,
> + .sync = FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT,
> + .vmode = FB_VMODE_NONINTERLACED,
> + .pixclock = 1,
> + .hsync_len = 6,
> + .vsync_len = 1,
> + .right_margin = 6,
> + .left_margin = 6,
> + },
> + .width = 71, /* 7.11cm */
> + .height = 53, /* 5.33cm */
> + .tim2 = 0x80007d0,
> + .cntl = CNTL_LCDMONO8,
> + .bpp = 8,
> + .grayscale = 1
> +};
Still no capabilities in the above - how does the CLCD controller know
which of BGR or RGB output and which format these panels support?
Each entry needs a .caps = CLCD_CAP_xxx | CLCD_CAP_xxx listing the
formats supported there.
Unless panels and the board have a .caps entry, the caps system won't
be used.
> +static struct clcd_board nspire_clcd_data = {
> + .name = "LCD",
> + .caps = CLCD_CAP_ALL,
Your board has logic to support the RGB444/BGR444 mode? If not, setting
CLCD_CAP_ALL is incorrect.
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 1/6] arm: Initial TI-Nspire support
@ 2013-05-12 9:06 ` Russell King - ARM Linux
0 siblings, 0 replies; 44+ messages in thread
From: Russell King - ARM Linux @ 2013-05-12 9:06 UTC (permalink / raw)
To: linux-arm-kernel
On Sun, May 12, 2013 at 02:22:56PM +1000, Daniel Tang wrote:
> diff --git a/arch/arm/mach-nspire/clcd.c b/arch/arm/mach-nspire/clcd.c
> new file mode 100644
> index 0000000..a4b9f06
> --- /dev/null
> +++ b/arch/arm/mach-nspire/clcd.c
> @@ -0,0 +1,117 @@
> +/*
> + * linux/arch/arm/mach-nspire/clcd.c
> + *
> + * Copyright (C) 2013 Daniel Tang <tangrs@tangrs.id.au>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2, as
> + * published by the Free Software Foundation.
> + *
> + */
> +
> +#include <linux/init.h>
> +#include <linux/of.h>
> +#include <linux/amba/bus.h>
> +#include <linux/amba/clcd.h>
> +#include <linux/dma-mapping.h>
> +
> +static struct clcd_panel nspire_cx_lcd_panel = {
> + .mode = {
> + .name = "Color LCD",
> + .refresh = 60,
> + .xres = 320,
> + .yres = 240,
> + .sync = 0,
> + .vmode = FB_VMODE_NONINTERLACED,
> + .pixclock = 1,
> + .hsync_len = 6,
> + .vsync_len = 1,
> + .right_margin = 50,
> + .left_margin = 38,
> + .lower_margin = 3,
> + .upper_margin = 17,
> + },
> + .width = 65, /* ~6.50 cm */
> + .height = 49, /* ~4.87 cm */
> + .tim2 = TIM2_IPC,
> + .cntl = CNTL_LCDTFT | CNTL_LCDVCOMP(1),
> + .bpp = 16,
> +};
> +
> +static struct clcd_panel nspire_classic_lcd_panel = {
> + .mode = {
> + .name = "Grayscale LCD",
> + .refresh = 60,
> + .xres = 320,
> + .yres = 240,
> + .sync = FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT,
> + .vmode = FB_VMODE_NONINTERLACED,
> + .pixclock = 1,
> + .hsync_len = 6,
> + .vsync_len = 1,
> + .right_margin = 6,
> + .left_margin = 6,
> + },
> + .width = 71, /* 7.11cm */
> + .height = 53, /* 5.33cm */
> + .tim2 = 0x80007d0,
> + .cntl = CNTL_LCDMONO8,
> + .bpp = 8,
> + .grayscale = 1
> +};
Still no capabilities in the above - how does the CLCD controller know
which of BGR or RGB output and which format these panels support?
Each entry needs a .caps = CLCD_CAP_xxx | CLCD_CAP_xxx listing the
formats supported there.
Unless panels and the board have a .caps entry, the caps system won't
be used.
> +static struct clcd_board nspire_clcd_data = {
> + .name = "LCD",
> + .caps = CLCD_CAP_ALL,
Your board has logic to support the RGB444/BGR444 mode? If not, setting
CLCD_CAP_ALL is incorrect.
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 1/6] arm: Initial TI-Nspire support
2013-05-12 9:06 ` Russell King - ARM Linux
@ 2013-05-12 9:22 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 9:22 UTC (permalink / raw)
To: Russell King - ARM Linux
Cc: linux-arm-kernel, Linus Walleij, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On 12/05/2013, at 7:06 PM, Russell King - ARM Linux <linux@arm.linux.org.uk> wrote:
> On Sun, May 12, 2013 at 02:22:56PM +1000, Daniel Tang wrote:
>>
>> + .bpp = 8,
>> + .grayscale = 1
>> +};
>
> Still no capabilities in the above - how does the CLCD controller know
> which of BGR or RGB output and which format these panels support?
>
> Each entry needs a .caps = CLCD_CAP_xxx | CLCD_CAP_xxx listing the
> formats supported there.
>
> Unless panels and the board have a .caps entry, the caps system won't
> be used.
Fair enough.
>
>> +static struct clcd_board nspire_clcd_data = {
>> + .name = "LCD",
>> + .caps = CLCD_CAP_ALL,
>
> Your board has logic to support the RGB444/BGR444 mode? If not, setting
> CLCD_CAP_ALL is incorrect.
I believe the grayscale panel supports the xGx444 modes - unsure about the colour one. Since I'm using the same struct clcd_board for both panels, I thought I'd just set caps to work with both to simplify.
Should I set board caps to panel caps on setup?
Cheers,
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 1/6] arm: Initial TI-Nspire support
@ 2013-05-12 9:22 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-12 9:22 UTC (permalink / raw)
To: linux-arm-kernel
On 12/05/2013, at 7:06 PM, Russell King - ARM Linux <linux@arm.linux.org.uk> wrote:
> On Sun, May 12, 2013 at 02:22:56PM +1000, Daniel Tang wrote:
>>
>> + .bpp = 8,
>> + .grayscale = 1
>> +};
>
> Still no capabilities in the above - how does the CLCD controller know
> which of BGR or RGB output and which format these panels support?
>
> Each entry needs a .caps = CLCD_CAP_xxx | CLCD_CAP_xxx listing the
> formats supported there.
>
> Unless panels and the board have a .caps entry, the caps system won't
> be used.
Fair enough.
>
>> +static struct clcd_board nspire_clcd_data = {
>> + .name = "LCD",
>> + .caps = CLCD_CAP_ALL,
>
> Your board has logic to support the RGB444/BGR444 mode? If not, setting
> CLCD_CAP_ALL is incorrect.
I believe the grayscale panel supports the xGx444 modes - unsure about the colour one. Since I'm using the same struct clcd_board for both panels, I thought I'd just set caps to work with both to simplify.
Should I set board caps to panel caps on setup?
Cheers,
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-14 8:03 ` Linus Walleij
-1 siblings, 0 replies; 44+ messages in thread
From: Linus Walleij @ 2013-05-14 8:03 UTC (permalink / raw)
To: Daniel Tang, Thomas Gleixner, John Stultz
Cc: linux-arm-kernel, Russell King - ARM Linux, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On Sun, May 12, 2013 at 6:22 AM, Daniel Tang <dt.tangr@gmail.com> wrote:
> Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
A commit message does not hurt.
> drivers/clocksource/Makefile | 1 +
> drivers/clocksource/nspire-classic-timer.c | 199 +++++++++++++++++++++++++++++
This subsystem is managed by Thomas Gleixner and John Stultz so you
should include
them on the To: line in reviewes.
> +++ b/drivers/clocksource/nspire-classic-timer.c
(...)
> +struct nspire_timer {
> + void __iomem *base;
> + void __iomem *timer1, *timer2;
> + void __iomem *interrupt_regs;
> +
> + int irqnr;
> +
> + struct clk *clk;
> + struct clock_event_device clkevt;
> + struct irqaction clkevt_irq;
> +
> + char clocksource_name[64];
> + char clockevent_name[64];
> +};
> +
> +static int nspire_timer_set_event(unsigned long delta,
> + struct clock_event_device *dev)
> +{
> + unsigned long flags;
> + struct nspire_timer *timer = container_of(dev,
> + struct nspire_timer,
> + clkevt);
> +
> + local_irq_save(flags);
> +
> + writel(delta, timer->timer1 + IO_CURRENT_VAL);
> + writel(CNTL_RUN_TIMER | CNTL_DEC | CNTL_MATCH1,
> + timer->timer1 + IO_CONTROL);
> +
> + local_irq_restore(flags);
> +
> + return 0;
> +}
> +static void nspire_timer_set_mode(enum clock_event_mode mode,
> + struct clock_event_device *evt)
> +{
> + evt->mode = mode;
So you support any mode?
Atleast handle CLOCK_EVT_MODE_SHUTDOWN and CLOCK_EVT_UNUSED
like any well-behaved clock event driver.
> +}
> +
> +static irqreturn_t nspire_timer_interrupt(int irq, void *dev_id)
> +{
> + struct nspire_timer *timer = dev_id;
> +
> + writel((1<<0), timer->interrupt_regs + IO_INTR_ACK);
write(0x1, timer->interrupt_regs + IO_INTR_ACK);
is no less intelligible, if you really want to point out that you're
setting bit 0, include <linux/bitops.h> and use BIT(0).
If there is a status bit you're clearing, maybe you should check
it first to watch for spurious IRQs?
> + writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
> +
> + if (timer->clkevt.event_handler)
> + timer->clkevt.event_handler(&timer->clkevt);
> +
> + return IRQ_HANDLED;
> +}
> +
> +static int __init nspire_timer_add(struct device_node *node)
> +{
> + struct nspire_timer *timer;
> + struct resource res;
> + int ret;
> +
> + timer = kzalloc(sizeof(*timer), GFP_ATOMIC);
> + if (!timer)
> + return -ENOMEM;
> +
> + timer->base = of_iomap(node, 0);
> + if (!timer->base) {
> + ret = -EINVAL;
> + goto error_free;
> + }
> + timer->timer1 = timer->base + IO_TIMER1;
> + timer->timer2 = timer->base + IO_TIMER2;
> +
> + timer->clk = of_clk_get(node, 0);
> + if (IS_ERR(timer->clk)) {
> + ret = PTR_ERR(timer->clk);
> + pr_err("Timer clock not found! (error %d)\n", ret);
> + goto error_unmap;
> + }
> +
> + timer->interrupt_regs = of_iomap(node, 1);
Check for errors.
> + timer->irqnr = irq_of_parse_and_map(node, 0);
Check for errors.
> +
> + of_address_to_resource(node, 0, &res);
> + scnprintf(timer->clocksource_name, sizeof(timer->clocksource_name),
> + "%llx.%s_clocksource",
> + (unsigned long long)res.start, node->name);
> +
> + scnprintf(timer->clockevent_name, sizeof(timer->clockevent_name),
> + "%llx.%s_clockevent",
> + (unsigned long long)res.start, node->name);
> +
> + if (timer->interrupt_regs && timer->irqnr) {
> + timer->clkevt.name = timer->clockevent_name;
> + timer->clkevt.set_next_event = nspire_timer_set_event;
> + timer->clkevt.set_mode = nspire_timer_set_mode;
> + timer->clkevt.rating = 200;
> + timer->clkevt.shift = 32;
Delete this shift. It will be set by clockevents_config_and_register().
> + timer->clkevt.cpumask = cpu_all_mask;
> + timer->clkevt.features =
> + CLOCK_EVT_FEAT_ONESHOT;
> +
> + writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
> + writel(0, timer->timer1 + IO_DIVIDER);
For a 16-bit counter I don't think it's wise to set the divider to 0
unless the clock frequency is very low. See further comments
below. (Assumes this is some prescaler.)
> + /* Mask out interrupts except the one we're using */
> + writel((1<<0), timer->interrupt_regs + IO_INTR_MSK);
Just 0x1 or BIT(0).
> + writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
Hm magic number, I guess it's clearing all IRQ lines...
> +
> + /* Interrupt to occur when timer value matches 0 */
> + writel(0, timer->base + IO_MATCH1);
> +
> + timer->clkevt_irq.name = timer->clockevent_name;
> + timer->clkevt_irq.handler = nspire_timer_interrupt;
> + timer->clkevt_irq.dev_id = timer;
> + timer->clkevt_irq.flags =
> + IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL;
> +
> + setup_irq(timer->irqnr, &timer->clkevt_irq);
> +
> + clockevents_config_and_register(&timer->clkevt,
> + clk_get_rate(timer->clk), 0x0001, 0xfffe);
> + pr_info("Added %s as clockevent\n", timer->clockevent_name);
> + }
> +
> + writel(CNTL_RUN_TIMER | CNTL_FOREVER | CNTL_INC,
> + timer->timer2 + IO_CONTROL);
> +
> + clocksource_mmio_init(timer->timer2 + IO_CURRENT_VAL,
> + timer->clocksource_name,
> + clk_get_rate(timer->clk),
> + 200, 16,
> + clocksource_mmio_readw_up);
If this timer is really just 16 bits, it's a *pretty* good idea to use
the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
and wake up unnecessarily often.
The same goes for the clock event.
See how we calculate the prescaler in arch/arm/mach-integrator/integrator_ap.c
for example.
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
@ 2013-05-14 8:03 ` Linus Walleij
0 siblings, 0 replies; 44+ messages in thread
From: Linus Walleij @ 2013-05-14 8:03 UTC (permalink / raw)
To: linux-arm-kernel
On Sun, May 12, 2013 at 6:22 AM, Daniel Tang <dt.tangr@gmail.com> wrote:
> Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
A commit message does not hurt.
> drivers/clocksource/Makefile | 1 +
> drivers/clocksource/nspire-classic-timer.c | 199 +++++++++++++++++++++++++++++
This subsystem is managed by Thomas Gleixner and John Stultz so you
should include
them on the To: line in reviewes.
> +++ b/drivers/clocksource/nspire-classic-timer.c
(...)
> +struct nspire_timer {
> + void __iomem *base;
> + void __iomem *timer1, *timer2;
> + void __iomem *interrupt_regs;
> +
> + int irqnr;
> +
> + struct clk *clk;
> + struct clock_event_device clkevt;
> + struct irqaction clkevt_irq;
> +
> + char clocksource_name[64];
> + char clockevent_name[64];
> +};
> +
> +static int nspire_timer_set_event(unsigned long delta,
> + struct clock_event_device *dev)
> +{
> + unsigned long flags;
> + struct nspire_timer *timer = container_of(dev,
> + struct nspire_timer,
> + clkevt);
> +
> + local_irq_save(flags);
> +
> + writel(delta, timer->timer1 + IO_CURRENT_VAL);
> + writel(CNTL_RUN_TIMER | CNTL_DEC | CNTL_MATCH1,
> + timer->timer1 + IO_CONTROL);
> +
> + local_irq_restore(flags);
> +
> + return 0;
> +}
> +static void nspire_timer_set_mode(enum clock_event_mode mode,
> + struct clock_event_device *evt)
> +{
> + evt->mode = mode;
So you support any mode?
Atleast handle CLOCK_EVT_MODE_SHUTDOWN and CLOCK_EVT_UNUSED
like any well-behaved clock event driver.
> +}
> +
> +static irqreturn_t nspire_timer_interrupt(int irq, void *dev_id)
> +{
> + struct nspire_timer *timer = dev_id;
> +
> + writel((1<<0), timer->interrupt_regs + IO_INTR_ACK);
write(0x1, timer->interrupt_regs + IO_INTR_ACK);
is no less intelligible, if you really want to point out that you're
setting bit 0, include <linux/bitops.h> and use BIT(0).
If there is a status bit you're clearing, maybe you should check
it first to watch for spurious IRQs?
> + writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
> +
> + if (timer->clkevt.event_handler)
> + timer->clkevt.event_handler(&timer->clkevt);
> +
> + return IRQ_HANDLED;
> +}
> +
> +static int __init nspire_timer_add(struct device_node *node)
> +{
> + struct nspire_timer *timer;
> + struct resource res;
> + int ret;
> +
> + timer = kzalloc(sizeof(*timer), GFP_ATOMIC);
> + if (!timer)
> + return -ENOMEM;
> +
> + timer->base = of_iomap(node, 0);
> + if (!timer->base) {
> + ret = -EINVAL;
> + goto error_free;
> + }
> + timer->timer1 = timer->base + IO_TIMER1;
> + timer->timer2 = timer->base + IO_TIMER2;
> +
> + timer->clk = of_clk_get(node, 0);
> + if (IS_ERR(timer->clk)) {
> + ret = PTR_ERR(timer->clk);
> + pr_err("Timer clock not found! (error %d)\n", ret);
> + goto error_unmap;
> + }
> +
> + timer->interrupt_regs = of_iomap(node, 1);
Check for errors.
> + timer->irqnr = irq_of_parse_and_map(node, 0);
Check for errors.
> +
> + of_address_to_resource(node, 0, &res);
> + scnprintf(timer->clocksource_name, sizeof(timer->clocksource_name),
> + "%llx.%s_clocksource",
> + (unsigned long long)res.start, node->name);
> +
> + scnprintf(timer->clockevent_name, sizeof(timer->clockevent_name),
> + "%llx.%s_clockevent",
> + (unsigned long long)res.start, node->name);
> +
> + if (timer->interrupt_regs && timer->irqnr) {
> + timer->clkevt.name = timer->clockevent_name;
> + timer->clkevt.set_next_event = nspire_timer_set_event;
> + timer->clkevt.set_mode = nspire_timer_set_mode;
> + timer->clkevt.rating = 200;
> + timer->clkevt.shift = 32;
Delete this shift. It will be set by clockevents_config_and_register().
> + timer->clkevt.cpumask = cpu_all_mask;
> + timer->clkevt.features =
> + CLOCK_EVT_FEAT_ONESHOT;
> +
> + writel(CNTL_STOP_TIMER, timer->timer1 + IO_CONTROL);
> + writel(0, timer->timer1 + IO_DIVIDER);
For a 16-bit counter I don't think it's wise to set the divider to 0
unless the clock frequency is very low. See further comments
below. (Assumes this is some prescaler.)
> + /* Mask out interrupts except the one we're using */
> + writel((1<<0), timer->interrupt_regs + IO_INTR_MSK);
Just 0x1 or BIT(0).
> + writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
Hm magic number, I guess it's clearing all IRQ lines...
> +
> + /* Interrupt to occur when timer value matches 0 */
> + writel(0, timer->base + IO_MATCH1);
> +
> + timer->clkevt_irq.name = timer->clockevent_name;
> + timer->clkevt_irq.handler = nspire_timer_interrupt;
> + timer->clkevt_irq.dev_id = timer;
> + timer->clkevt_irq.flags =
> + IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL;
> +
> + setup_irq(timer->irqnr, &timer->clkevt_irq);
> +
> + clockevents_config_and_register(&timer->clkevt,
> + clk_get_rate(timer->clk), 0x0001, 0xfffe);
> + pr_info("Added %s as clockevent\n", timer->clockevent_name);
> + }
> +
> + writel(CNTL_RUN_TIMER | CNTL_FOREVER | CNTL_INC,
> + timer->timer2 + IO_CONTROL);
> +
> + clocksource_mmio_init(timer->timer2 + IO_CURRENT_VAL,
> + timer->clocksource_name,
> + clk_get_rate(timer->clk),
> + 200, 16,
> + clocksource_mmio_readw_up);
If this timer is really just 16 bits, it's a *pretty* good idea to use
the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
and wake up unnecessarily often.
The same goes for the clock event.
See how we calculate the prescaler in arch/arm/mach-integrator/integrator_ap.c
for example.
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
2013-05-14 8:03 ` Linus Walleij
@ 2013-05-14 11:51 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-14 11:51 UTC (permalink / raw)
To: Linus Walleij
Cc: Thomas Gleixner, John Stultz, linux-arm-kernel,
Russell King - ARM Linux, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On 14/05/2013, at 6:03 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>> +
>> + timer->interrupt_regs = of_iomap(node, 1);
>
> Check for errors.
>
>> + timer->irqnr = irq_of_parse_and_map(node, 0);
>
> Check for errors.
>> + if (timer->interrupt_regs && timer->irqnr) {
>>
The idea with this is to make these properties optional. Each timer has two sub-timers but only one can generate interrupts. One will be added as a clockevent device and the other as a clocksource. If an IRQ number or interrupt acknowledge address is not passed through the DT, then the clockevents device is not added but the clocksource still is.
Should I comment it in the code to make it apparent what's happening or is there a better way to express this?
>
>> + writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
>
> Hm magic number, I guess it's clearing all IRQ lines…
It's a bitmask of bits 0-5.
>
>> +
>> + /* Interrupt to occur when timer value matches 0 */
>> + writel(0, timer->base + IO_MATCH1);
>> +
>> + timer->clkevt_irq.name = timer->clockevent_name;
>> + timer->clkevt_irq.handler = nspire_timer_interrupt;
>> + timer->clkevt_irq.dev_id = timer;
>> + timer->clkevt_irq.flags =
>> + IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL;
>> +
>> + setup_irq(timer->irqnr, &timer->clkevt_irq);
>> +
>> + clockevents_config_and_register(&timer->clkevt,
>> + clk_get_rate(timer->clk), 0x0001, 0xfffe);
>> + pr_info("Added %s as clockevent\n", timer->clockevent_name);
>> + }
>> +
>> + writel(CNTL_RUN_TIMER | CNTL_FOREVER | CNTL_INC,
>> + timer->timer2 + IO_CONTROL);
>> +
>> + clocksource_mmio_init(timer->timer2 + IO_CURRENT_VAL,
>> + timer->clocksource_name,
>> + clk_get_rate(timer->clk),
>> + 200, 16,
>> + clocksource_mmio_readw_up);
>
> If this timer is really just 16 bits, it's a *pretty* good idea to use
> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
> and wake up unnecessarily often.
>
> The same goes for the clock event.
The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
>
> See how we calculate the prescaler in arch/arm/mach-integrator/integrator_ap.c
> for example.
>
> Yours,
> Linus Walleij
Cheers,
Daniel Tang
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
@ 2013-05-14 11:51 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-14 11:51 UTC (permalink / raw)
To: linux-arm-kernel
On 14/05/2013, at 6:03 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>> +
>> + timer->interrupt_regs = of_iomap(node, 1);
>
> Check for errors.
>
>> + timer->irqnr = irq_of_parse_and_map(node, 0);
>
> Check for errors.
>> + if (timer->interrupt_regs && timer->irqnr) {
>>
The idea with this is to make these properties optional. Each timer has two sub-timers but only one can generate interrupts. One will be added as a clockevent device and the other as a clocksource. If an IRQ number or interrupt acknowledge address is not passed through the DT, then the clockevents device is not added but the clocksource still is.
Should I comment it in the code to make it apparent what's happening or is there a better way to express this?
>
>> + writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
>
> Hm magic number, I guess it's clearing all IRQ lines?
It's a bitmask of bits 0-5.
>
>> +
>> + /* Interrupt to occur when timer value matches 0 */
>> + writel(0, timer->base + IO_MATCH1);
>> +
>> + timer->clkevt_irq.name = timer->clockevent_name;
>> + timer->clkevt_irq.handler = nspire_timer_interrupt;
>> + timer->clkevt_irq.dev_id = timer;
>> + timer->clkevt_irq.flags =
>> + IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL;
>> +
>> + setup_irq(timer->irqnr, &timer->clkevt_irq);
>> +
>> + clockevents_config_and_register(&timer->clkevt,
>> + clk_get_rate(timer->clk), 0x0001, 0xfffe);
>> + pr_info("Added %s as clockevent\n", timer->clockevent_name);
>> + }
>> +
>> + writel(CNTL_RUN_TIMER | CNTL_FOREVER | CNTL_INC,
>> + timer->timer2 + IO_CONTROL);
>> +
>> + clocksource_mmio_init(timer->timer2 + IO_CURRENT_VAL,
>> + timer->clocksource_name,
>> + clk_get_rate(timer->clk),
>> + 200, 16,
>> + clocksource_mmio_readw_up);
>
> If this timer is really just 16 bits, it's a *pretty* good idea to use
> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
> and wake up unnecessarily often.
>
> The same goes for the clock event.
The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
>
> See how we calculate the prescaler in arch/arm/mach-integrator/integrator_ap.c
> for example.
>
> Yours,
> Linus Walleij
Cheers,
Daniel Tang
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-15 14:07 ` Arnd Bergmann
-1 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-15 14:07 UTC (permalink / raw)
To: Daniel Tang
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On Sunday 12 May 2013, Daniel Tang wrote:
>
> Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
> ---
> drivers/clk/Makefile | 1 +
> drivers/clk/clk-nspire.c | 141 +++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 142 insertions(+)
> create mode 100644 drivers/clk/clk-nspire.c
You are missing a binding in Documentation/devicetree, same as for some of
the other drivers in this series.
> +static int nspire_clk_read(struct device_node *node,
> + struct nspire_clk_info *clk)
> +{
> + u32 val;
> + int ret;
> + void __iomem *io;
> + const char *type = NULL;
> +
> + ret = of_property_read_string(node, "io-type", &type);
> + if (ret)
> + return ret;
Using a string here feels clumsy. Why not use the "compatible" property
to distinguish the different models?
> +static void __init nspire_ahbdiv_setup(struct device_node *node)
> +{
> + int ret;
> + struct clk *clk;
> + const char *clk_name = node->name;
> + const char *parent_name;
> + struct nspire_clk_info info;
> +
> + ret = nspire_clk_read(node, &info);
> + if (WARN_ON(ret))
> + return;
> +
> + of_property_read_string(node, "clock-output-names", &clk_name);
It seems strange to assign the clk_name variable to node->name
first and then overriding it with the clock-output-names property.
Is that intentional? If so, please explain it in a comment.
> +
> + pr_info("TI-NSPIRE Base: %uMHz CPU: %uMHz AHB: %uMHz\n",
> + info.base_clock / MHZ,
> + info.base_clock / info.base_cpu_ratio / MHZ,
> + info.base_clock / info.base_ahb_ratio / MHZ);
> +}
> +
> +CLK_OF_DECLARE(nspire_clk, "nspire-clock", nspire_clk_setup);
> +CLK_OF_DECLARE(nspire_ahbdiv, "nspire-ahb-divider", nspire_ahbdiv_setup);
I would put each of these lines directly under the function it references.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-15 14:07 ` Arnd Bergmann
0 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-15 14:07 UTC (permalink / raw)
To: linux-arm-kernel
On Sunday 12 May 2013, Daniel Tang wrote:
>
> Signed-off-by: Daniel Tang <dt.tangr@gmail.com>
> ---
> drivers/clk/Makefile | 1 +
> drivers/clk/clk-nspire.c | 141 +++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 142 insertions(+)
> create mode 100644 drivers/clk/clk-nspire.c
You are missing a binding in Documentation/devicetree, same as for some of
the other drivers in this series.
> +static int nspire_clk_read(struct device_node *node,
> + struct nspire_clk_info *clk)
> +{
> + u32 val;
> + int ret;
> + void __iomem *io;
> + const char *type = NULL;
> +
> + ret = of_property_read_string(node, "io-type", &type);
> + if (ret)
> + return ret;
Using a string here feels clumsy. Why not use the "compatible" property
to distinguish the different models?
> +static void __init nspire_ahbdiv_setup(struct device_node *node)
> +{
> + int ret;
> + struct clk *clk;
> + const char *clk_name = node->name;
> + const char *parent_name;
> + struct nspire_clk_info info;
> +
> + ret = nspire_clk_read(node, &info);
> + if (WARN_ON(ret))
> + return;
> +
> + of_property_read_string(node, "clock-output-names", &clk_name);
It seems strange to assign the clk_name variable to node->name
first and then overriding it with the clock-output-names property.
Is that intentional? If so, please explain it in a comment.
> +
> + pr_info("TI-NSPIRE Base: %uMHz CPU: %uMHz AHB: %uMHz\n",
> + info.base_clock / MHZ,
> + info.base_clock / info.base_cpu_ratio / MHZ,
> + info.base_clock / info.base_ahb_ratio / MHZ);
> +}
> +
> +CLK_OF_DECLARE(nspire_clk, "nspire-clock", nspire_clk_setup);
> +CLK_OF_DECLARE(nspire_ahbdiv, "nspire-ahb-divider", nspire_ahbdiv_setup);
I would put each of these lines directly under the function it references.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 2/6] arm: Add device trees for TI-Nspire
2013-05-12 4:22 ` Daniel Tang
@ 2013-05-15 14:10 ` Arnd Bergmann
-1 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-15 14:10 UTC (permalink / raw)
To: Daniel Tang
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On Sunday 12 May 2013, Daniel Tang wrote:
> +
> + timer0: timer@900C0000 {
> + reg = <0x900C0000 0x1000>;
> +
> + clocks = <&timer_clk>;
> + clock-names = "timer_clk";
> + };
> +
> + timer1: timer@900D0000 {
> + reg = <0x900D0000 0x1000>;
> + interrupts = <19>;
> +
> + clocks = <&timer_clk>;
> + clock-names = "timer_clk";
> + };
> + clocks = <&apb_pclk>;
> + clock-names = "apb_pclk";
> + };
Why do you provide "clock-names" here? Do those devices have multiple
clock inputs? It's normally easier to just use the default clock
input.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 2/6] arm: Add device trees for TI-Nspire
@ 2013-05-15 14:10 ` Arnd Bergmann
0 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-15 14:10 UTC (permalink / raw)
To: linux-arm-kernel
On Sunday 12 May 2013, Daniel Tang wrote:
> +
> + timer0: timer at 900C0000 {
> + reg = <0x900C0000 0x1000>;
> +
> + clocks = <&timer_clk>;
> + clock-names = "timer_clk";
> + };
> +
> + timer1: timer at 900D0000 {
> + reg = <0x900D0000 0x1000>;
> + interrupts = <19>;
> +
> + clocks = <&timer_clk>;
> + clock-names = "timer_clk";
> + };
> + clocks = <&apb_pclk>;
> + clock-names = "apb_pclk";
> + };
Why do you provide "clock-names" here? Do those devices have multiple
clock inputs? It's normally easier to just use the default clock
input.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-15 14:07 ` Arnd Bergmann
@ 2013-05-16 8:31 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-16 8:31 UTC (permalink / raw)
To: Arnd Bergmann
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On 16/05/2013, at 12:07 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> You are missing a binding in Documentation/devicetree, same as for some of
> the other drivers in this series.
Should we be adding a vendor prefix to it too? If so, we're not sure whether to use "ti," or not since this isn't an official port by TI.
> It seems strange to assign the clk_name variable to node->name
> first and then overriding it with the clock-output-names property.
> Is that intentional? If so, please explain it in a comment.
>
I copied that bit of boilerplate from drivers/clk/clk-fixed-rate.c but I'm guessing it's to use the node name as the clock name unless there is a property called "clock-output-names"
Cheers,
Daniel Tang
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-16 8:31 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-16 8:31 UTC (permalink / raw)
To: linux-arm-kernel
On 16/05/2013, at 12:07 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> You are missing a binding in Documentation/devicetree, same as for some of
> the other drivers in this series.
Should we be adding a vendor prefix to it too? If so, we're not sure whether to use "ti," or not since this isn't an official port by TI.
> It seems strange to assign the clk_name variable to node->name
> first and then overriding it with the clock-output-names property.
> Is that intentional? If so, please explain it in a comment.
>
I copied that bit of boilerplate from drivers/clk/clk-fixed-rate.c but I'm guessing it's to use the node name as the clock name unless there is a property called "clock-output-names"
Cheers,
Daniel Tang
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-16 8:31 ` Daniel Tang
@ 2013-05-16 12:17 ` Arnd Bergmann
-1 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-16 12:17 UTC (permalink / raw)
To: Daniel Tang
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel,
Mike Turquette
On Thursday 16 May 2013, Daniel Tang wrote:
>
> On 16/05/2013, at 12:07 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>
> > You are missing a binding in Documentation/devicetree, same as for some of
> > the other drivers in this series.
>
> Should we be adding a vendor prefix to it too? If so, we're not sure whether
> to use "ti," or not since this isn't an official port by TI.
The binding describes the hardware, it should not matter who does the port.
However, any part of the tree that is not actually from TI should have a
vendor prefix indicating who made that part. IIRC, the SoC used in there
is from TI, so you should use something else for the on-soc components.
> > It seems strange to assign the clk_name variable to node->name
> > first and then overriding it with the clock-output-names property.
> > Is that intentional? If so, please explain it in a comment.
> >
>
> I copied that bit of boilerplate from drivers/clk/clk-fixed-rate.c but
> I'm guessing it's to use the node name as the clock name unless there
> is a property called "clock-output-names"
Ah, I see. It seems you forgot to add the clock maintainer to Cc in the
mail. Mike is the one who will have to take you patch anyway, so I assume
he will comment on this if you did it wrong.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-16 12:17 ` Arnd Bergmann
0 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-16 12:17 UTC (permalink / raw)
To: linux-arm-kernel
On Thursday 16 May 2013, Daniel Tang wrote:
>
> On 16/05/2013, at 12:07 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>
> > You are missing a binding in Documentation/devicetree, same as for some of
> > the other drivers in this series.
>
> Should we be adding a vendor prefix to it too? If so, we're not sure whether
> to use "ti," or not since this isn't an official port by TI.
The binding describes the hardware, it should not matter who does the port.
However, any part of the tree that is not actually from TI should have a
vendor prefix indicating who made that part. IIRC, the SoC used in there
is from TI, so you should use something else for the on-soc components.
> > It seems strange to assign the clk_name variable to node->name
> > first and then overriding it with the clock-output-names property.
> > Is that intentional? If so, please explain it in a comment.
> >
>
> I copied that bit of boilerplate from drivers/clk/clk-fixed-rate.c but
> I'm guessing it's to use the node name as the clock name unless there
> is a property called "clock-output-names"
Ah, I see. It seems you forgot to add the clock maintainer to Cc in the
mail. Mike is the one who will have to take you patch anyway, so I assume
he will comment on this if you did it wrong.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
2013-05-14 11:51 ` Daniel Tang
@ 2013-05-17 13:17 ` Linus Walleij
-1 siblings, 0 replies; 44+ messages in thread
From: Linus Walleij @ 2013-05-17 13:17 UTC (permalink / raw)
To: Daniel Tang
Cc: Thomas Gleixner, John Stultz, linux-arm-kernel,
Russell King - ARM Linux, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On Tue, May 14, 2013 at 1:51 PM, Daniel Tang <dt.tangr@gmail.com> wrote:
> On 14/05/2013, at 6:03 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>
>>> +
>>> + timer->interrupt_regs = of_iomap(node, 1);
>>
>> Check for errors.
>>
>>> + timer->irqnr = irq_of_parse_and_map(node, 0);
>>
>> Check for errors.
>>> + if (timer->interrupt_regs && timer->irqnr) {
>>>
>
> The idea with this is to make these properties optional. Each timer has two sub-timers
> but only one can generate interrupts.
I see now. Sorry, I was just being dumb, don't worry!
> Should I comment it in the code to make it apparent what's happening or is there a
> better way to express this?
Maybe. Maybe it's obvious to everyone else but me :-/
>>> + writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
>>
>> Hm magic number, I guess it's clearing all IRQ lines…
>
> It's a bitmask of bits 0-5.
Yeah I figured so much... So 6 IRQ lines that get cleared.
>> If this timer is really just 16 bits, it's a *pretty* good idea to use
>> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
>> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
>> and wake up unnecessarily often.
>>
>> The same goes for the clock event.
>
> The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
No. But that was *pretty* slow.
The thing is that the driver is getting "some clockfrequency" from
the clk subsystem and does not know what it is.
Usually (and I've seen this hardware design pattern a lot) when you
have a clock like this, and go talk to the clk implementation specialist
s/he will say "oh, yeah, I can mux in this 34 MHz clock to the timer
as well", and if that is possible then that is often what you want,
because it's good for things like HRtimers, and then it's useful to have
a prescaler och the clocksource/clockevent.
You can use that 32768 Hz for clock events and clocksource (but I
suspect that HRtimers will never be useful on this platform due to this).
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
@ 2013-05-17 13:17 ` Linus Walleij
0 siblings, 0 replies; 44+ messages in thread
From: Linus Walleij @ 2013-05-17 13:17 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, May 14, 2013 at 1:51 PM, Daniel Tang <dt.tangr@gmail.com> wrote:
> On 14/05/2013, at 6:03 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>
>>> +
>>> + timer->interrupt_regs = of_iomap(node, 1);
>>
>> Check for errors.
>>
>>> + timer->irqnr = irq_of_parse_and_map(node, 0);
>>
>> Check for errors.
>>> + if (timer->interrupt_regs && timer->irqnr) {
>>>
>
> The idea with this is to make these properties optional. Each timer has two sub-timers
> but only one can generate interrupts.
I see now. Sorry, I was just being dumb, don't worry!
> Should I comment it in the code to make it apparent what's happening or is there a
> better way to express this?
Maybe. Maybe it's obvious to everyone else but me :-/
>>> + writel(0x3F, timer->interrupt_regs + IO_INTR_ACK);
>>
>> Hm magic number, I guess it's clearing all IRQ lines?
>
> It's a bitmask of bits 0-5.
Yeah I figured so much... So 6 IRQ lines that get cleared.
>> If this timer is really just 16 bits, it's a *pretty* good idea to use
>> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
>> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
>> and wake up unnecessarily often.
>>
>> The same goes for the clock event.
>
> The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
No. But that was *pretty* slow.
The thing is that the driver is getting "some clockfrequency" from
the clk subsystem and does not know what it is.
Usually (and I've seen this hardware design pattern a lot) when you
have a clock like this, and go talk to the clk implementation specialist
s/he will say "oh, yeah, I can mux in this 34 MHz clock to the timer
as well", and if that is possible then that is often what you want,
because it's good for things like HRtimers, and then it's useful to have
a prescaler och the clocksource/clockevent.
You can use that 32768 Hz for clock events and clocksource (but I
suspect that HRtimers will never be useful on this platform due to this).
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
2013-05-17 13:17 ` Linus Walleij
@ 2013-05-18 6:40 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-18 6:40 UTC (permalink / raw)
To: Linus Walleij
Cc: Thomas Gleixner, John Stultz, linux-arm-kernel,
Russell King - ARM Linux, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On 17/05/2013, at 11:17 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>>> If this timer is really just 16 bits, it's a *pretty* good idea to use
>>> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
>>> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
>>> and wake up unnecessarily often.
>>>
>>> The same goes for the clock event.
>>
>> The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
>
> No. But that was *pretty* slow.
>
> The thing is that the driver is getting "some clockfrequency" from
> the clk subsystem and does not know what it is.
>
> Usually (and I've seen this hardware design pattern a lot) when you
> have a clock like this, and go talk to the clk implementation specialist
> s/he will say "oh, yeah, I can mux in this 34 MHz clock to the timer
> as well", and if that is possible then that is often what you want,
> because it's good for things like HRtimers, and then it's useful to have
> a prescaler och the clocksource/clockevent.
>
> You can use that 32768 Hz for clock events and clocksource (but I
> suspect that HRtimers will never be useful on this platform due to this).
>
I forgot to mention before that there is also another timer that runs at the same freq as the APB (which is usually around 30MHz). Is this what you were alluding to? I haven't tested the driver on that timer yet.
As I understand it, you're saying to use a prescaler on faster clock rates so the kernel can have a longer maximum delay time (since a 16 bit register can only hold only so much). In that case, what kind of range should I be scaling to? I suspect it needs to be large enough to keep the kernel from constantly waking but small enough for HR timers to be useful.
> Yours,
> Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
@ 2013-05-18 6:40 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-18 6:40 UTC (permalink / raw)
To: linux-arm-kernel
On 17/05/2013, at 11:17 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>>> If this timer is really just 16 bits, it's a *pretty* good idea to use
>>> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
>>> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
>>> and wake up unnecessarily often.
>>>
>>> The same goes for the clock event.
>>
>> The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
>
> No. But that was *pretty* slow.
>
> The thing is that the driver is getting "some clockfrequency" from
> the clk subsystem and does not know what it is.
>
> Usually (and I've seen this hardware design pattern a lot) when you
> have a clock like this, and go talk to the clk implementation specialist
> s/he will say "oh, yeah, I can mux in this 34 MHz clock to the timer
> as well", and if that is possible then that is often what you want,
> because it's good for things like HRtimers, and then it's useful to have
> a prescaler och the clocksource/clockevent.
>
> You can use that 32768 Hz for clock events and clocksource (but I
> suspect that HRtimers will never be useful on this platform due to this).
>
I forgot to mention before that there is also another timer that runs at the same freq as the APB (which is usually around 30MHz). Is this what you were alluding to? I haven't tested the driver on that timer yet.
As I understand it, you're saying to use a prescaler on faster clock rates so the kernel can have a longer maximum delay time (since a 16 bit register can only hold only so much). In that case, what kind of range should I be scaling to? I suspect it needs to be large enough to keep the kernel from constantly waking but small enough for HR timers to be useful.
> Yours,
> Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-16 12:17 ` Arnd Bergmann
@ 2013-05-19 11:09 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-19 11:09 UTC (permalink / raw)
To: Arnd Bergmann
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel,
Mike Turquette
On 16/05/2013, at 10:17 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Thursday 16 May 2013, Daniel Tang wrote:
>>
>> On 16/05/2013, at 12:07 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>>
>>> You are missing a binding in Documentation/devicetree, same as for some of
>>> the other drivers in this series.
>>
>> Should we be adding a vendor prefix to it too? If so, we're not sure whether
>> to use "ti," or not since this isn't an official port by TI.
>
> The binding describes the hardware, it should not matter who does the port.
> However, any part of the tree that is not actually from TI should have a
> vendor prefix indicating who made that part. IIRC, the SoC used in there
> is from TI, so you should use something else for the on-soc components.
>
If the vendors for the on-SOC components are unknown, should we just leave the compatible strings as is (i.e. "nspire-XXX")?
>>> It seems strange to assign the clk_name variable to node->name
>>> first and then overriding it with the clock-output-names property.
>>> Is that intentional? If so, please explain it in a comment.
>>>
>>
>> I copied that bit of boilerplate from drivers/clk/clk-fixed-rate.c but
>> I'm guessing it's to use the node name as the clock name unless there
>> is a property called "clock-output-names"
>
> Ah, I see. It seems you forgot to add the clock maintainer to Cc in the
> mail. Mike is the one who will have to take you patch anyway, so I assume
> he will comment on this if you did it wrong.
>
> Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-19 11:09 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-19 11:09 UTC (permalink / raw)
To: linux-arm-kernel
On 16/05/2013, at 10:17 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Thursday 16 May 2013, Daniel Tang wrote:
>>
>> On 16/05/2013, at 12:07 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>>
>>> You are missing a binding in Documentation/devicetree, same as for some of
>>> the other drivers in this series.
>>
>> Should we be adding a vendor prefix to it too? If so, we're not sure whether
>> to use "ti," or not since this isn't an official port by TI.
>
> The binding describes the hardware, it should not matter who does the port.
> However, any part of the tree that is not actually from TI should have a
> vendor prefix indicating who made that part. IIRC, the SoC used in there
> is from TI, so you should use something else for the on-soc components.
>
If the vendors for the on-SOC components are unknown, should we just leave the compatible strings as is (i.e. "nspire-XXX")?
>>> It seems strange to assign the clk_name variable to node->name
>>> first and then overriding it with the clock-output-names property.
>>> Is that intentional? If so, please explain it in a comment.
>>>
>>
>> I copied that bit of boilerplate from drivers/clk/clk-fixed-rate.c but
>> I'm guessing it's to use the node name as the clock name unless there
>> is a property called "clock-output-names"
>
> Ah, I see. It seems you forgot to add the clock maintainer to Cc in the
> mail. Mike is the one who will have to take you patch anyway, so I assume
> he will comment on this if you did it wrong.
>
> Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-19 11:09 ` Daniel Tang
@ 2013-05-19 19:48 ` Arnd Bergmann
-1 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-19 19:48 UTC (permalink / raw)
To: Daniel Tang
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel,
Mike Turquette
On Sunday 19 May 2013, Daniel Tang wrote:
> If the vendors for the on-SOC components are unknown, should we just
> leave the compatible strings as is (i.e. "nspire-XXX")?
In that case, I would use the name of the company that made the SoC.
I believe someone mentioned it was made by LSI logic.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-19 19:48 ` Arnd Bergmann
0 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-19 19:48 UTC (permalink / raw)
To: linux-arm-kernel
On Sunday 19 May 2013, Daniel Tang wrote:
> If the vendors for the on-SOC components are unknown, should we just
> leave the compatible strings as is (i.e. "nspire-XXX")?
In that case, I would use the name of the company that made the SoC.
I believe someone mentioned it was made by LSI logic.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-19 19:48 ` Arnd Bergmann
@ 2013-05-20 11:19 ` Daniel Tang
-1 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-20 11:19 UTC (permalink / raw)
To: Arnd Bergmann
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel,
Mike Turquette
On 20/05/2013, at 5:48 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Sunday 19 May 2013, Daniel Tang wrote:
>> If the vendors for the on-SOC components are unknown, should we just
>> leave the compatible strings as is (i.e. "nspire-XXX")?
>
> In that case, I would use the name of the company that made the SoC.
> I believe someone mentioned it was made by LSI logic.
Yep, that's right. So "lsi,nspire-XXX" should be fine?
>
> Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-20 11:19 ` Daniel Tang
0 siblings, 0 replies; 44+ messages in thread
From: Daniel Tang @ 2013-05-20 11:19 UTC (permalink / raw)
To: linux-arm-kernel
On 20/05/2013, at 5:48 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Sunday 19 May 2013, Daniel Tang wrote:
>> If the vendors for the on-SOC components are unknown, should we just
>> leave the compatible strings as is (i.e. "nspire-XXX")?
>
> In that case, I would use the name of the company that made the SoC.
> I believe someone mentioned it was made by LSI logic.
Yep, that's right. So "lsi,nspire-XXX" should be fine?
>
> Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
2013-05-20 11:19 ` Daniel Tang
@ 2013-05-20 12:26 ` Arnd Bergmann
-1 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-20 12:26 UTC (permalink / raw)
To: Daniel Tang
Cc: linux-arm-kernel, linux, Linus Walleij,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel,
Mike Turquette
On Monday 20 May 2013 21:19:33 Daniel Tang wrote:
> On 20/05/2013, at 5:48 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>
> > On Sunday 19 May 2013, Daniel Tang wrote:
> >> If the vendors for the on-SOC components are unknown, should we just
> >> leave the compatible strings as is (i.e. "nspire-XXX")?
> >
> > In that case, I would use the name of the company that made the SoC.
> > I believe someone mentioned it was made by LSI logic.
>
> Yep, that's right. So "lsi,nspire-XXX" should be fine?
I wouldn't use that combination, since lsi did not make the nspire.
For anything on the LSI chip, "lsi,zevio-XXX" would be correct,
or maybe "lsi,zevio1020-XXX" for something specific to a particular
SoC variant. For off-SoC parts, "ti,nspire-XXX" would be right,
or "ti,nspire-cx-XXX" if you care about the model.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers
@ 2013-05-20 12:26 ` Arnd Bergmann
0 siblings, 0 replies; 44+ messages in thread
From: Arnd Bergmann @ 2013-05-20 12:26 UTC (permalink / raw)
To: linux-arm-kernel
On Monday 20 May 2013 21:19:33 Daniel Tang wrote:
> On 20/05/2013, at 5:48 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>
> > On Sunday 19 May 2013, Daniel Tang wrote:
> >> If the vendors for the on-SOC components are unknown, should we just
> >> leave the compatible strings as is (i.e. "nspire-XXX")?
> >
> > In that case, I would use the name of the company that made the SoC.
> > I believe someone mentioned it was made by LSI logic.
>
> Yep, that's right. So "lsi,nspire-XXX" should be fine?
I wouldn't use that combination, since lsi did not make the nspire.
For anything on the LSI chip, "lsi,zevio-XXX" would be correct,
or maybe "lsi,zevio1020-XXX" for something specific to a particular
SoC variant. For off-SoC parts, "ti,nspire-XXX" would be right,
or "ti,nspire-cx-XXX" if you care about the model.
Arnd
^ permalink raw reply [flat|nested] 44+ messages in thread
* Re: [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
2013-05-18 6:40 ` Daniel Tang
@ 2013-05-20 12:49 ` Linus Walleij
-1 siblings, 0 replies; 44+ messages in thread
From: Linus Walleij @ 2013-05-20 12:49 UTC (permalink / raw)
To: Daniel Tang
Cc: Thomas Gleixner, John Stultz, linux-arm-kernel,
Russell King - ARM Linux, Arnd Bergmann,
fabian@ritter-vogt.de Vogt, Lionel Debroux, linux-kernel
On Sat, May 18, 2013 at 8:40 AM, Daniel Tang <dt.tangr@gmail.com> wrote:
> On 17/05/2013, at 11:17 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>
>>>> If this timer is really just 16 bits, it's a *pretty* good idea to use
>>>> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
>>>> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
>>>> and wake up unnecessarily often.
>>>>
>>>> The same goes for the clock event.
>>>
>>> The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
>>
>> No. But that was *pretty* slow.
>>
>> The thing is that the driver is getting "some clockfrequency" from
>> the clk subsystem and does not know what it is.
>>
>> Usually (and I've seen this hardware design pattern a lot) when you
>> have a clock like this, and go talk to the clk implementation specialist
>> s/he will say "oh, yeah, I can mux in this 34 MHz clock to the timer
>> as well", and if that is possible then that is often what you want,
>> because it's good for things like HRtimers, and then it's useful to have
>> a prescaler och the clocksource/clockevent.
>>
>> You can use that 32768 Hz for clock events and clocksource (but I
>> suspect that HRtimers will never be useful on this platform due to this).
>>
>
> I forgot to mention before that there is also another timer that runs
> at the same freq as the APB (which is usually around 30MHz).
> Is this what you were alluding to? I haven't tested the driver on that timer yet.
OK then that is probably the timer you will use for clock event and
also sched_clock() in the end. (You'll be setting a better rating on it.)
I'm not alluding to it as I didn't know of it, but it makes HW-sense
that it exists :-)
> As I understand it, you're saying to use a prescaler on faster clock rates
> so the kernel can have a longer maximum delay time (since a 16 bit register
> can only hold only so much). In that case, what kind of range should I be
> scaling to? I suspect it needs to be large enough to keep the kernel from
> constantly waking but small enough for HR timers to be useful.
Yeah :-/
This is one of the cases where one thing excludes the other.
But if it's hard-wired to 32kHz you probably want to leave it as it
is.
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
* [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers
@ 2013-05-20 12:49 ` Linus Walleij
0 siblings, 0 replies; 44+ messages in thread
From: Linus Walleij @ 2013-05-20 12:49 UTC (permalink / raw)
To: linux-arm-kernel
On Sat, May 18, 2013 at 8:40 AM, Daniel Tang <dt.tangr@gmail.com> wrote:
> On 17/05/2013, at 11:17 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
>
>>>> If this timer is really just 16 bits, it's a *pretty* good idea to use
>>>> the prescaler (I guess this is what IO_DIVIDER is) beacuse else you
>>>> will get short sleep times with CONFIG_NO_HZ_IDLE on this system,
>>>> and wake up unnecessarily often.
>>>>
>>>> The same goes for the clock event.
>>>
>>> The clock frequency is 32768Hz. Should I be scaling it down at that frequency?
>>
>> No. But that was *pretty* slow.
>>
>> The thing is that the driver is getting "some clockfrequency" from
>> the clk subsystem and does not know what it is.
>>
>> Usually (and I've seen this hardware design pattern a lot) when you
>> have a clock like this, and go talk to the clk implementation specialist
>> s/he will say "oh, yeah, I can mux in this 34 MHz clock to the timer
>> as well", and if that is possible then that is often what you want,
>> because it's good for things like HRtimers, and then it's useful to have
>> a prescaler och the clocksource/clockevent.
>>
>> You can use that 32768 Hz for clock events and clocksource (but I
>> suspect that HRtimers will never be useful on this platform due to this).
>>
>
> I forgot to mention before that there is also another timer that runs
> at the same freq as the APB (which is usually around 30MHz).
> Is this what you were alluding to? I haven't tested the driver on that timer yet.
OK then that is probably the timer you will use for clock event and
also sched_clock() in the end. (You'll be setting a better rating on it.)
I'm not alluding to it as I didn't know of it, but it makes HW-sense
that it exists :-)
> As I understand it, you're saying to use a prescaler on faster clock rates
> so the kernel can have a longer maximum delay time (since a 16 bit register
> can only hold only so much). In that case, what kind of range should I be
> scaling to? I suspect it needs to be large enough to keep the kernel from
> constantly waking but small enough for HR timers to be useful.
Yeah :-/
This is one of the cases where one thing excludes the other.
But if it's hard-wired to 32kHz you probably want to leave it as it
is.
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 44+ messages in thread
end of thread, other threads:[~2013-05-20 12:49 UTC | newest]
Thread overview: 44+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-05-12 4:22 [RFC PATCHv3 0/6] arm: Initial TI-Nspire support Daniel Tang
2013-05-12 4:22 ` Daniel Tang
2013-05-12 4:22 ` [RFC PATCHv3 1/6] " Daniel Tang
2013-05-12 4:22 ` Daniel Tang
2013-05-12 9:06 ` Russell King - ARM Linux
2013-05-12 9:06 ` Russell King - ARM Linux
2013-05-12 9:22 ` Daniel Tang
2013-05-12 9:22 ` Daniel Tang
2013-05-12 4:22 ` [RFC PATCHv3 2/6] arm: Add device trees for TI-Nspire Daniel Tang
2013-05-12 4:22 ` Daniel Tang
2013-05-15 14:10 ` Arnd Bergmann
2013-05-15 14:10 ` Arnd Bergmann
2013-05-12 4:22 ` [RFC PATCHv3 3/6] clk: Add TI-Nspire clock drivers Daniel Tang
2013-05-12 4:22 ` Daniel Tang
2013-05-15 14:07 ` Arnd Bergmann
2013-05-15 14:07 ` Arnd Bergmann
2013-05-16 8:31 ` Daniel Tang
2013-05-16 8:31 ` Daniel Tang
2013-05-16 12:17 ` Arnd Bergmann
2013-05-16 12:17 ` Arnd Bergmann
2013-05-19 11:09 ` Daniel Tang
2013-05-19 11:09 ` Daniel Tang
2013-05-19 19:48 ` Arnd Bergmann
2013-05-19 19:48 ` Arnd Bergmann
2013-05-20 11:19 ` Daniel Tang
2013-05-20 11:19 ` Daniel Tang
2013-05-20 12:26 ` Arnd Bergmann
2013-05-20 12:26 ` Arnd Bergmann
2013-05-12 4:22 ` [RFC PATCHv3 4/6] clocksource: Add TI-Nspire timer drivers Daniel Tang
2013-05-12 4:22 ` Daniel Tang
2013-05-14 8:03 ` Linus Walleij
2013-05-14 8:03 ` Linus Walleij
2013-05-14 11:51 ` Daniel Tang
2013-05-14 11:51 ` Daniel Tang
2013-05-17 13:17 ` Linus Walleij
2013-05-17 13:17 ` Linus Walleij
2013-05-18 6:40 ` Daniel Tang
2013-05-18 6:40 ` Daniel Tang
2013-05-20 12:49 ` Linus Walleij
2013-05-20 12:49 ` Linus Walleij
2013-05-12 4:23 ` [RFC PATCHv3 5/6] input: Add TI-Nspire keypad driver Daniel Tang
2013-05-12 4:23 ` Daniel Tang
2013-05-12 4:23 ` [RFC PATCHv3 6/6] irqchip: Add TI-Nspire irqchip Daniel Tang
2013-05-12 4:23 ` Daniel Tang
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.