Skip to contentSkip to conversation

dt-bindings: timer: Add binding doc for nxp system counter timer

PATCH Superseded Jacky Bai 3 participants 4 messages Apr 3, 2019 – Apr 6, 2019

Subsystems (3)

v3 of 5 Superseded Latest is v5 →

lore
Share & tools
lore

Jacky Bai

author
#0 lore raw

PATCH v3 1/2 dt-bindings: timer: Add binding doc for nxp system counter timer

From: Bai Ping <[email protected]>

Add the binding doc for nxp system counter timer module.

---
change v1->v2
- remove the blank line at EOF
change v2->v3
- update the binding example based on the driver change

---
 .../devicetree/bindings/timer/nxp,sysctr-timer.txt | 25 ++++++++++++++++++++++
 1 file changed, 25 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/timer/nxp,sysctr-timer.txt
Documentation/devicetree/bindings/timer/nxp,sysctr-timer.txt new file +25 -0
new file mode 100644
index 0000000..d576599
@@ -0,0 +1,25 @@
+NXP System Counter Module(sys_ctr)
+
+The system counter(sys_ctr) is a programmable system counter which provides
+a shared time base to Cortex A15, A7, A53, A73, etc. it is intended for use in
+applications where the counter is always powered and support multiple,
+unrelated clocks. The compare frame inside can be used for timer purpose.
+
+Required properties:
+
+- compatible :      should be "nxp,sysctr-timer"
+- reg :             Specifies the base physical address and size of the comapre
+                    frame and the counter control, read & compare.
+- interrupts :      should be the first compare frames' interrupt
+- clocks : 	    Specifies the counter clock.
+- clock-names: 	    Specifies the clock's name of this module
+
+Example:
+
+	system_counter: timer@306a0000 {
+		compatible = "nxp,sysctr-timer";
+		reg = <0x306a0000 0x20000>;/* system-counter-rd & compare */
+		clocks = <&clk_8m>;
+		clock-names = "per";
+		interrupts = <GIC_SPI 47 IRQ_TYPE_LEVEL_HIGH>;
+	};
Signature · 1 line
1.9.1
author
#1 lore raw

PATCH v3 2/2 driver: clocksource: Add nxp system counter timer driver support

From: Bai Ping <[email protected]>

The system counter (sys_ctr) is a programmable system counter which provides a shared time base to the Cortex A15, A7, A53 etc cores. It is intended for use in applications where the counter is always powered on and supports multiple, unrelated clocks. The sys_ctr hardware supports:
- 56-bit counter width (roll-over time greater than 40 years)
- compare frame(64-bit compare value) contains programmable interrupt generation

---
change v1->v2:
- no change
change v2->v3:
- remove the clocksource, we only need to use this module for timer purpose, so register it as clockevent is enough.
- use the timer_of_init to init the irq, clock, etc.
- remove some unnecessary comments.

---
 drivers/clocksource/Kconfig            |   7 ++
 drivers/clocksource/Makefile           |   1 +
 drivers/clocksource/timer-imx-sysctr.c | 149 +++++++++++++++++++++++++++++++++
 3 files changed, 157 insertions(+)
 create mode 100644 drivers/clocksource/timer-imx-sysctr.c
drivers/clocksource/Kconfig +7 -0
index 171502a..90cd908 100644
@@ -598,6 +598,13 @@ config CLKSRC_IMX_TPM
 	  Enable this option to use IMX Timer/PWM Module (TPM) timer as
 	  clocksource.
 
+config TIMER_IMX_SYS_CTR
+	bool "i.MX system counter timer" if COMPILE_TEST
+	depends on ARCH_MXC
+	select TIMER_OF
+	help
+	  Enable this option to use i.MX system counter timer for clockevent.
+
 config CLKSRC_ST_LPC
 	bool "Low power clocksource found in the LPC" if COMPILE_TEST
 	select TIMER_OF if OF
drivers/clocksource/Makefile +1 -0
index be6e0fb..bc5ce50 100644
@@ -73,6 +73,7 @@ obj-$(CONFIG_CLKSRC_MIPS_GIC)		+= mips-gic-timer.o
 obj-$(CONFIG_CLKSRC_TANGO_XTAL)		+= timer-tango-xtal.o
 obj-$(CONFIG_CLKSRC_IMX_GPT)		+= timer-imx-gpt.o
 obj-$(CONFIG_CLKSRC_IMX_TPM)		+= timer-imx-tpm.o
+obj-$(CONFIG_TIMER_IMX_SYS_CTR)		+= timer-imx-sysctr.o
 obj-$(CONFIG_ASM9260_TIMER)		+= asm9260_timer.o
 obj-$(CONFIG_H8300_TMR8)		+= h8300_timer8.o
 obj-$(CONFIG_H8300_TMR16)		+= h8300_timer16.o
drivers/clocksource/timer-imx-sysctr.c new file +149 -0

149 lines not on this page · raw message

Signature · 1 line
1.9.1
#2 lore raw

Re: [PATCH v3 2/2] driver: clocksource: Add nxp system counter timer driver support

Quoted 10 lines · Jacky Bai

From: Bai Ping <[email protected]>

The system counter (sys_ctr) is a programmable system counter which provides a shared time base to the Cortex A15, A7, A53 etc cores. It is intended for use in applications where the counter is always powered on and supports multiple, unrelated clocks. The sys_ctr hardware supports:
- 56-bit counter width (roll-over time greater than 40 years)
- compare frame(64-bit compare value) contains programmable interrupt generation

I hope that's a <= compare and not a == ....

drivers/clocksource/timer-imx-sysctr.c line 34

Quoted 10 lines
drivers/clocksource/timer-imx-sysctr.c new file +10 -0
+static void sysctr_timer_enable(bool enable)
+{
+	u32 val;
+
+	val = readl(sys_ctr_base + CMPCR);
+	val &= ~SYS_CTR_EN;
+	if (enable)
+		val |= SYS_CTR_EN;
+
+	writel(val, sys_ctr_base + CMPCR);

This read is really just overhead. Why aren't you caching the control register value? It's not a self modifying register and I don't see concurrency here either.

Quoted 11 lines
drivers/clocksource/timer-imx-sysctr.c new file +11 -0
+}
+
+static void sysctr_irq_acknowledge(void)
+{
+	/*
+	 * clear the enable bit(EN =0) will clear
+	 * the status bit(ISTAT = 0), then the interrupt
+	 * signal will be negated(acknowledged).
+	 */
+	sysctr_timer_enable(false);
+}

drivers/clocksource/timer-imx-sysctr.c line 55

Quoted 10 lines
drivers/clocksource/timer-imx-sysctr.c new file +10 -0
+
+static inline u64 sysctr_read_counter(void)
+{
+	u32 cnt_hi, tmp_hi, cnt_lo;
+
+	do {
+		cnt_hi = readl_relaxed(sys_ctr_base + CNTCV_HI);
+		cnt_lo = readl_relaxed(sys_ctr_base + CNTCV_LO);
+		tmp_hi = readl_relaxed(sys_ctr_base + CNTCV_HI);
+	} while (tmp_hi != cnt_hi);

When will hardware people finally get it? Is it so damned hard to make the readout do:

	lo = read_lo()		-> internally latches HI in hardware
	hi = read_hi()		-> reads the latched value

It's not rocket science, but it would spare these horrible read loops. But sure, performance happens in whitepapers and marketing slides ....

Quoted 11 lines
drivers/clocksource/timer-imx-sysctr.c new file +11 -0
+
+	return  ((u64) cnt_hi << 32) | cnt_lo;
+}
+
+static int sysctr_set_next_event(unsigned long delta,
+				 struct clock_event_device *evt)
+{
+	u32 cmp_hi, cmp_lo;
+	u64 next;
+
+	sysctr_timer_enable(false);

drivers/clocksource/timer-imx-sysctr.c line 76

Quoted 10 lines
drivers/clocksource/timer-imx-sysctr.c new file +10 -0
+
+	next = sysctr_read_counter();
+
+	next += delta;
+
+	cmp_hi = (next >> 32) & 0x00fffff;
+	cmp_lo = next & 0xffffffff;
+
+	writel_relaxed(cmp_hi, sys_ctr_base + CMPCV_HI);
+	writel_relaxed(cmp_lo, sys_ctr_base + CMPCV_LO);

Please document that this is a <= comparator. If that's not true then this function is broken for small deltas and delays between read_counter() and enable.

drivers/clocksource/timer-imx-sysctr.c line 85

Quoted 9 lines
drivers/clocksource/timer-imx-sysctr.c new file +9 -0
+
+	sysctr_timer_enable(true);
+
+	return 0;
+}
+
+static int sysctr_set_state_oneshot(struct clock_event_device *evt)
+{
+	sysctr_timer_enable(true);

That's wrong. Why do you want to enable the timer here? When the state is set to one shot then the next operation is set_next_event() but before that nothing should ever come out of the timer.

Thanks,

tglx

maintainer Reviewed-by on v3 1/2
#3 lore raw
In reply to Jacky Bai
Quoted 10 lines · Jacky Bai

From: Bai Ping <[email protected]>

Add the binding doc for nxp system counter timer module.

---
change v1->v2
- remove the blank line at EOF
change v2->v3
- update the binding example based on the driver change

Quoted 5 lines
---
 .../devicetree/bindings/timer/nxp,sysctr-timer.txt | 25 ++++++++++++++++++++++
 1 file changed, 25 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/timer/nxp,sysctr-timer.txt