lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <VI1PR0402MB3519447E5135AC3C040758F387810@VI1PR0402MB3519.eurprd04.prod.outlook.com>
Date:   Tue, 15 Jan 2019 02:49:30 +0000
From:   Jacky Bai <ping.bai@....com>
To:     Daniel Lezcano <daniel.lezcano@...aro.org>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        "robh+dt@...nel.org" <robh+dt@...nel.org>,
        "shawnguo@...nel.org" <shawnguo@...nel.org>,
        "mark.rutland@....com" <mark.rutland@....com>,
        Aisheng Dong <aisheng.dong@....com>
CC:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        dl-linux-imx <linux-imx@....com>
Subject: RE: [PATCH v2 2/2] driver: clocksource: Add nxp system counter timer
 driver support

> Subject: Re: [PATCH v2 2/2] driver: clocksource: Add nxp system counter timer
> driver support
> 
> On 12/12/2018 06:28, Jacky Bai wrote:
> > 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
> >
> > Signed-off-by: Bai Ping <ping.bai@....com>
> > ---
> > change v1->v2:
> >  - no change
> > ---
> >  drivers/clocksource/Kconfig            |   8 ++
> >  drivers/clocksource/Makefile           |   1 +
> >  drivers/clocksource/timer-imx-sysctr.c | 204
> > +++++++++++++++++++++++++++++++++
> >  3 files changed, 213 insertions(+)
> >  create mode 100644 drivers/clocksource/timer-imx-sysctr.c
> >
> > diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
> > index c57b156..7e5f2de 100644
> > --- a/drivers/clocksource/Kconfig
> > +++ b/drivers/clocksource/Kconfig
> > @@ -579,6 +579,14 @@ config CLKSRC_IMX_TPM
> >  	  Enable this option to use IMX Timer/PWM Module (TPM) timer as
> >  	  clocksource.
> >
> > +config CLKSRC_IMX_SYS_CTR
> 
> As the driver is a timer, please replace the CLKSRC by TIMER

Ok, will fix it.

> 
> > +	bool "Clocksource using i.MX system counter" if COMPILE_TEST
> > +	depends on (ARM || ARM64) && CLKDEV_LOOKUP
> 
> CLKDEV_LOOKUP is not needed, it is selected with COMMON_CLK
> 
> > +	select CLKSRC_MMIO
> > +	help
> > +	  Enable this option to use IMX system counter timer as
> > +	  clocksource.
> > +
> >  config CLKSRC_ST_LPC
> >  	bool "Low power clocksource found in the LPC" if COMPILE_TEST
> >  	select TIMER_OF if OF
> > diff --git a/drivers/clocksource/Makefile
> > b/drivers/clocksource/Makefile index dd91381..372bf4e 100644
> > --- a/drivers/clocksource/Makefile
> > +++ b/drivers/clocksource/Makefile
> > @@ -71,6 +71,7 @@ obj-$(CONFIG_CLKSRC_MIPS_GIC)		+=
> mips-gic-timer.o
> >  obj-$(CONFIG_CLKSRC_TANGO_XTAL)		+= tango_xtal.o
> >  obj-$(CONFIG_CLKSRC_IMX_GPT)		+= timer-imx-gpt.o
> >  obj-$(CONFIG_CLKSRC_IMX_TPM)		+= timer-imx-tpm.o
> > +obj-$(CONFIG_CLKSRC_IMX_SYS_CTR)	+= timer-imx-sysctr.o
> >  obj-$(CONFIG_ASM9260_TIMER)		+= asm9260_timer.o
> >  obj-$(CONFIG_H8300_TMR8)		+= h8300_timer8.o
> >  obj-$(CONFIG_H8300_TMR16)		+= h8300_timer16.o
> > diff --git a/drivers/clocksource/timer-imx-sysctr.c
> > b/drivers/clocksource/timer-imx-sysctr.c
> > new file mode 100644
> > index 0000000..ad3c27f
> > --- /dev/null
> > +++ b/drivers/clocksource/timer-imx-sysctr.c
> > @@ -0,0 +1,204 @@
> > +// SPDX-License-Identifier: GPL-2.0+
> > +//
> > +// Copyright 2017-2018 NXP
> > +
> > +#include <linux/interrupt.h>
> > +#include <linux/clockchips.h>
> > +#include <linux/clocksource.h>
> > +#include <linux/delay.h>
> > +#include <linux/of_address.h>
> > +#include <linux/of_irq.h>
> > +#include <linux/sched_clock.h>
> > +
> > +#define CNTCV_LO	0x8
> > +#define CNTCV_HI	0xc
> > +#define CMPCV_LO	0x20
> > +#define CMPCV_HI	0x24
> > +#define CMPCR		0x2c
> > +
> > +#define SYS_CTR_EN		0x1
> > +#define SYS_CTR_IRQ_MASK	0x2
> > +
> > +static void __iomem *sys_ctr_rd_base; static void __iomem
> > +*sys_ctr_cmp_base; static struct clock_event_device
> > +clockevent_sysctr;
> > +
> > +static inline void sysctr_timer_enable(bool enable)
> 
> Remove 'inline' tag
> 

OK.

> > +{
> > +	u32 val;
> > +
> > +	val = readl(sys_ctr_cmp_base + CMPCR);
> > +	val &= ~SYS_CTR_EN;
> > +	if (enable)
> > +		val |= SYS_CTR_EN;
> > +
> > +	writel(val, sys_ctr_cmp_base + CMPCR); }
> > +
> > +static void sysctr_irq_acknowledge(void) {
> > +	u32 val;
> > +
> > +	/* clear th enable bit(EN=0) to clear the ISTAT */
> 
> What is the ISTAT?
> 
> Thanks for adding comments, however the format is:
> 
> /*
>  *
>  */
> 

Th ISTAT is interrupt status bit. I will fix it.

> > +	val = readl(sys_ctr_cmp_base + CMPCR);
> > +	val &= ~SYS_CTR_EN;
> > +	writel(val, sys_ctr_cmp_base + CMPCR);
> 
> This is duplicate with the sysctr_timer_enable() function, call it instead of
> duplicating code.
> 
Sure.

> > +}
> > +
> > +static inline u64 sysctr_read_counter(void) {
> > +	u32 cnt_hi, tmp_hi, cnt_lo;
> > +
> > +	do {
> > +		cnt_hi = readl_relaxed(sys_ctr_rd_base + CNTCV_HI);
> > +		cnt_lo = readl_relaxed(sys_ctr_rd_base + CNTCV_LO);
> > +		tmp_hi = readl_relaxed(sys_ctr_rd_base + CNTCV_HI);
> > +	} while (tmp_hi != cnt_hi);
> > +
> > +	return  ((u64) cnt_hi << 32) | cnt_lo; }
> 
> It has been showed a 64bit counter usage adds a non negligible overhead,
> especially on 32bits. Are you really sure you want this ?
> 

This HW is designed as 56 bit counter & timer. when doing timer 'set_next_event',
The 56bit counter value must be read out, and then adding the delta value to set the
next event. Can you give me some suggestion on how to handle this?

> > +static u64 notrace sysctr_read_sched_clock(void) {
> > +	return sysctr_read_counter();
> > +}
> > +
> > +static u64 sysctr_clocksource_read(struct clocksource *cs) {
> > +	return sysctr_read_counter();
> > +}
> > +
> > +static int __init sysctr_clocksource_init(unsigned int rate) {
> > +	sched_clock_register(sysctr_read_sched_clock, 56, rate);
> > +	return clocksource_mmio_init(sys_ctr_rd_base, "i.MX sys_ctr",
> > +				     rate, 200, 56, sysctr_clocksource_read); }
> > +
> > +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);
> > +
> > +	next = sysctr_read_counter();
> > +
> > +	next += delta;
> > +
> > +	cmp_hi = (next >> 32) & 0x00fffff;
> > +	cmp_lo = next & 0xffffffff;
> > +
> > +	writel_relaxed(cmp_hi, sys_ctr_cmp_base + CMPCV_HI);
> > +	writel_relaxed(cmp_lo, sys_ctr_cmp_base + CMPCV_LO);
> > +
> > +	sysctr_timer_enable(true);
> > +
> > +	return 0;
> > +}
> > +
> > +static int sysctr_set_state_oneshot(struct clock_event_device *evt) {
> > +	/* enable timer */
> 
> The function name is explicit enough, no need to add this comment
> 

OK, will remove it in V3.

> > +	sysctr_timer_enable(true);
> > +
> > +	return 0;
> > +}
> > +
> > +static int sysctr_set_state_shutdown(struct clock_event_device *evt)
> > +{
> > +	/* disable the timer */
> 
> Ditto
> 
> > +	sysctr_timer_enable(false);
> > +
> > +	return 0;
> > +}
> > +
> > +static irqreturn_t sysctr_timer_interrupt(int irq, void *dev_id) {
> > +	struct clock_event_device *evt = &clockevent_sysctr;
> > +
> > +	sysctr_irq_acknowledge();
> > +
> > +	evt->event_handler(evt);
> > +
> > +	return IRQ_HANDLED;
> > +}
> > +
> > +static struct clock_event_device clockevent_sysctr = {
> > +	.name			= "i.MX system counter timer",
> > +	.features		= CLOCK_EVT_FEAT_ONESHOT |
> CLOCK_EVT_FEAT_DYNIRQ,
> > +	.set_state_oneshot	= sysctr_set_state_oneshot,
> > +	.set_next_event		= sysctr_set_next_event,
> > +	.set_state_shutdown	= sysctr_set_state_shutdown,
> > +	.rating			= 200,
> > +};
> > +
> > +static int __init sysctr_clockevent_init(unsigned long rate, int irq)
> > +{
> > +	int ret;
> > +
> > +	ret = request_irq(irq, sysctr_timer_interrupt, IRQF_TIMER |
> IRQF_IRQPOLL,
> > +		    "i.MX system counter timer", &clockevent_sysctr);
> > +	if (ret) {
> > +		pr_err("Failed to request i.MX sysctr timer irq\n");
> > +		return ret;
> > +	}
> > +
> > +	clockevent_sysctr.cpumask = cpumask_of(0);
> > +	clockevent_sysctr.irq = irq;
> > +	clockevents_config_and_register(&clockevent_sysctr,
> > +			rate, 0xff, 0x7fffffff);
> > +	return 0;
> > +}
> > +
> > +static int __init sysctr_timer_init(struct device_node *np) {
> > +	u32 rate;
> > +	int irq, ret = 0;
> > +
> > +	/* map the system counter's CNTreadbase */
> 
> Comment not useful, remove it
> 
> > +	sys_ctr_rd_base = of_iomap(np, 0);
> > +	if (!sys_ctr_rd_base) {
> > +		pr_err("Failed to map sys_ctr rd base%pOF\n", np);
> 
> nit: space missing after 'base'
> 
> > +		return -ENXIO;
> > +	}
> > +
> > +	/* map the system counter's CNTcomparebase */
> 
> Remove comment
> 
> > +	sys_ctr_cmp_base = of_iomap(np, 1);
> > +	if (!sys_ctr_cmp_base) {
> > +		pr_err("Failed to map sys_ctr compare base%pOF\n", np);
> > +		ret = -ENXIO;
> > +		goto out_free2;
> > +	}
> > +
> > +	/*
> > +	 * the purpose of this driver is to provide a global timer,
> > +	 * So only use one compare frame, request frame0's irq only.
> > +	 */
> > +	irq = irq_of_parse_and_map(np, 0);
> > +	if (!irq) {
> > +		pr_err("Failed to map interrupt for %pOF\n", np);
> > +		ret = -EINVAL;
> > +		goto out_free1;
> > +	}
> > +
> > +	if (of_property_read_u32(np, "clock-frequency", &rate)) {
> > +		pr_err("Failed to get clock frequency %pOF\n", np);
> > +		ret = -EINVAL;
> > +		goto out_free1;
> > +	}
> 
> Please replace it with a 'fixed-clock' in the DT and use the clk API.
> 
> In addition you can use the timer-of API (cf timer-of.h)
> 

Thanks, I will fix it in V3.

> > +
> > +	sysctr_clocksource_init(rate);
> > +	sysctr_clockevent_init(rate, irq);
> > +
> > +	return 0;
> > +
> > +out_free1:
> 
> label names are inadequate, use out_iounmap_cmp: for example
> 

Will fix it in V3.

BR

> > +	iounmap(sys_ctr_cmp_base);
> > +out_free2:
> > +	iounmap(sys_ctr_rd_base);
> > +	return ret;
> > +}
> > +TIMER_OF_DECLARE(sysctr_timer, "nxp,sysctr-timer",
> > +sysctr_timer_init);
> >

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ