linux-watchdog.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Paul Cercueil <paul@crapouillou.net>
To: Mathieu Malaterre <malat@debian.org>
Cc: Thierry Reding <thierry.reding@gmail.com>,
	Rob Herring <robh+dt@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ralf Baechle <ralf@linux-mips.org>,
	Paul Burton <paul.burton@mips.com>,
	James Hogan <jhogan@kernel.org>, Jonathan Corbet <corbet@lwn.net>,
	Ezequiel Garcia <ezequiel@collabora.co.uk>,
	PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>,
	linux-pwm@vger.kernel.org,
	"open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS" 
	<devicetree@vger.kernel.org>, LKML <linux-kernel@vger.kernel.org>,
	linux-watchdog@vger.kernel.org, linux-mips@vger.kernel.org,
	linux-doc@vger.kernel.org, linux-clk@vger.kernel.org, od@zcrc.me,
	Maarten ter Huurne <maarten@treewalker.org>
Subject: Re: [PATCH v8 05/26] clocksource: Add driver for the Ingenic JZ47xx OST
Date: Wed, 23 Jan 2019 14:27:25 -0300	[thread overview]
Message-ID: <1548264445.3173.2@crapouillou.net> (raw)
In-Reply-To: <CA+7wUsxNN2SL9_7tDh0DiBAgdUwxd_osgmi=09HP7110Tm0DYQ@mail.gmail.com>

(re-send, in plain text this time, sorry for those who got it twice)


Le mer. 23 janv. 2019 à 9:58, Mathieu Malaterre <malat@debian.org> a 
écrit :
> On Wed, Dec 12, 2018 at 11:09 PM Paul Cercueil <paul@crapouillou.net> 
> wrote:
>> 
>>  From: Maarten ter Huurne <maarten@treewalker.org>
>> 
>>  OST is the OS Timer, a 64-bit timer/counter with buffered reading.
>> 
>>  SoCs before the JZ4770 had (if any) a 32-bit OST; the JZ4770 and
>>  JZ4780 have a 64-bit OST.
>> 
>>  This driver will register both a clocksource and a sched_clock to 
>> the
>>  system.
>> 
>>  Signed-off-by: Maarten ter Huurne <maarten@treewalker.org>
>>  Signed-off-by: Paul Cercueil <paul@crapouillou.net>
>>  ---
>> 
>>  Notes:
>>       v5: New patch
>> 
>>       v6: - Get rid of SoC IDs; pass pointer to ingenic_ost_soc_info 
>> as
>>             devicetree match data instead.
>>           - Use device_get_match_data() instead of the of_* variant
>>           - Handle error of dev_get_regmap() properly
>> 
>>       v7: Fix section mismatch by using 
>> builtin_platform_driver_probe()
>> 
>>       v8: builtin_platform_driver_probe() does not work anymore in
>>           4.20-rc6? The probe function won't be called. Work around 
>> this
>>           for now by using late_initcall.
>> 
>>   drivers/clocksource/Kconfig       |   8 ++
>>   drivers/clocksource/Makefile      |   1 +
>>   drivers/clocksource/ingenic-ost.c | 215 
>> ++++++++++++++++++++++++++++++++++++++
>>   3 files changed, 224 insertions(+)
>>   create mode 100644 drivers/clocksource/ingenic-ost.c
>> 
>>  diff --git a/drivers/clocksource/Kconfig 
>> b/drivers/clocksource/Kconfig
>>  index 4e69af15c3e7..e0646878b0de 100644
>>  --- a/drivers/clocksource/Kconfig
>>  +++ b/drivers/clocksource/Kconfig
>>  @@ -648,4 +648,12 @@ config INGENIC_TIMER
>>          help
>>            Support for the timer/counter unit of the Ingenic JZ SoCs.
>> 
>>  +config INGENIC_OST
>>  +       bool "Ingenic JZ47xx Operating System Timer"
>>  +       depends on MIPS || COMPILE_TEST
>>  +       depends on COMMON_CLK
>>  +       select INGENIC_TIMER
>>  +       help
>>  +         Support for the OS Timer of the Ingenic JZ4770 or similar 
>> SoC.
>>  +
>>   endmenu
>>  diff --git a/drivers/clocksource/Makefile 
>> b/drivers/clocksource/Makefile
>>  index 7c8f790dcf67..7faa8108574a 100644
>>  --- a/drivers/clocksource/Makefile
>>  +++ b/drivers/clocksource/Makefile
>>  @@ -75,6 +75,7 @@ obj-$(CONFIG_ASM9260_TIMER)           += 
>> asm9260_timer.o
>>   obj-$(CONFIG_H8300_TMR8)               += h8300_timer8.o
>>   obj-$(CONFIG_H8300_TMR16)              += h8300_timer16.o
>>   obj-$(CONFIG_H8300_TPU)                        += h8300_tpu.o
>>  +obj-$(CONFIG_INGENIC_OST)              += ingenic-ost.o
>>   obj-$(CONFIG_INGENIC_TIMER)            += ingenic-timer.o
>>   obj-$(CONFIG_CLKSRC_ST_LPC)            += clksrc_st_lpc.o
>>   obj-$(CONFIG_X86_NUMACHIP)             += numachip.o
>>  diff --git a/drivers/clocksource/ingenic-ost.c 
>> b/drivers/clocksource/ingenic-ost.c
>>  new file mode 100644
>>  index 000000000000..cc0fee3efd29
>>  --- /dev/null
>>  +++ b/drivers/clocksource/ingenic-ost.c
>>  @@ -0,0 +1,215 @@
>>  +// SPDX-License-Identifier: GPL-2.0
>>  +/*
>>  + * JZ47xx SoCs TCU Operating System Timer driver
>>  + *
>>  + * Copyright (C) 2016 Maarten ter Huurne <maarten@treewalker.org>
>>  + * Copyright (C) 2018 Paul Cercueil <paul@crapouillou.net>
>>  + */
>>  +
>>  +#include <linux/clk.h>
>>  +#include <linux/clocksource.h>
>>  +#include <linux/mfd/ingenic-tcu.h>
>>  +#include <linux/of.h>
>>  +#include <linux/platform_device.h>
>>  +#include <linux/pm.h>
>>  +#include <linux/regmap.h>
>>  +#include <linux/sched_clock.h>
>>  +
>>  +#include "ingenic-timer.h"
>>  +
>>  +#define TCU_OST_TCSR_MASK      0xc0
>>  +#define TCU_OST_TCSR_CNT_MD    BIT(15)
>>  +
>>  +#define TCU_OST_CHANNEL                15
>>  +
>>  +struct ingenic_ost_soc_info {
>>  +       bool is64bit;
>>  +};
>>  +
>>  +struct ingenic_ost {
>>  +       struct regmap *map;
>>  +       struct clk *clk;
>>  +
>>  +       struct clocksource cs;
>>  +};
>>  +
>>  +static u64 notrace ingenic_ost_read_cntl(void)
>>  +{
>>  +       /* Bypass the regmap here as we must return as soon as 
>> possible */
>>  +       return readl(ingenic_tcu_base + TCU_REG_OST_CNTL);
>>  +}
>>  +
>>  +static u64 notrace ingenic_ost_read_cnth(void)
>>  +{
>>  +       /* Bypass the regmap here as we must return as soon as 
>> possible */
>>  +       return readl(ingenic_tcu_base + TCU_REG_OST_CNTH);
>>  +}
>>  +
>>  +static u64 notrace ingenic_ost_clocksource_read(struct clocksource 
>> *cs)
>>  +{
>>  +       u32 val1, val2;
>>  +       u64 count, recount;
>>  +       s64 diff;
>>  +
>>  +       /*
>>  +        * The buffering of the upper 32 bits of the timer prevents 
>> wrong
>>  +        * results from the bottom 32 bits overflowing due to the 
>> timer ticking
>>  +        * along. However, it does not prevent wrong results from 
>> simultaneous
>>  +        * reads of the timer, which could reset the buffer 
>> mid-read.
>>  +        * Since this kind of wrong read can happen only when the 
>> bottom bits
>>  +        * overflow, there will be minutes between wrong reads, so 
>> if we read
>>  +        * twice in succession, at least one of the reads will be 
>> correct.
>>  +        */
>>  +
>>  +       /* Bypass the regmap here as we must return as soon as 
>> possible */
>>  +       val1 = readl(ingenic_tcu_base + TCU_REG_OST_CNTL);
>>  +       val2 = readl(ingenic_tcu_base + TCU_REG_OST_CNTHBUF);
>>  +       count = (u64)val1 | (u64)val2 << 32;
>>  +
>>  +       val1 = readl(ingenic_tcu_base + TCU_REG_OST_CNTL);
>>  +       val2 = readl(ingenic_tcu_base + TCU_REG_OST_CNTHBUF);
>>  +       recount = (u64)val1 | (u64)val2 << 32;
>>  +
>>  +       /*
>>  +        * A wrong read will produce a result that is 1<<32 too 
>> high: the bottom
>>  +        * part from before overflow and the upper part from after 
>> overflow.
>>  +        * Therefore, the lower value of the two reads is the 
>> correct value.
>>  +        */
>>  +
>>  +       diff = (s64)(recount - count);
>>  +       if (unlikely(diff < 0))
>>  +               count = recount;
>>  +
>>  +       return count;
>>  +}
>>  +
>>  +static int __init ingenic_ost_probe(struct platform_device *pdev)
>>  +{
>>  +       const struct ingenic_ost_soc_info *soc_info;
>>  +       struct device *dev = &pdev->dev;
>>  +       struct ingenic_ost *ost;
>>  +       struct clocksource *cs;
>>  +       unsigned long rate, flags;
>>  +       int err;
>>  +
>>  +       soc_info = device_get_match_data(dev);
>>  +       if (!soc_info)
>>  +               return -EINVAL;
>>  +
>>  +       ost = devm_kzalloc(dev, sizeof(*ost), GFP_KERNEL);
>>  +       if (!ost)
>>  +               return -ENOMEM;
>>  +
>>  +       ost->map = dev_get_regmap(dev->parent, NULL);
>>  +       if (!ost->map) {
>>  +               dev_err(dev, "regmap not found\n");
>>  +               return -EINVAL;
>>  +       }
>>  +
>>  +       ost->clk = devm_clk_get(dev, "ost");
>>  +       if (IS_ERR(ost->clk))
>>  +               return PTR_ERR(ost->clk);
>>  +
>>  +       err = clk_prepare_enable(ost->clk);
>>  +       if (err)
>>  +               return err;
>>  +
>>  +       /* Clear counter high/low registers */
>>  +       if (soc_info->is64bit)
>>  +               regmap_write(ost->map, TCU_REG_OST_CNTL, 0);
>>  +       regmap_write(ost->map, TCU_REG_OST_CNTH, 0);
>>  +
>>  +       /* Don't reset counter at compare value. */
>>  +       regmap_update_bits(ost->map, TCU_REG_OST_TCSR,
>>  +                          TCU_OST_TCSR_MASK, TCU_OST_TCSR_CNT_MD);
>>  +
>>  +       rate = clk_get_rate(ost->clk);
>>  +
>>  +       /* Enable OST TCU channel */
>>  +       regmap_write(ost->map, TCU_REG_TESR, BIT(TCU_OST_CHANNEL));
>>  +
>>  +       cs = &ost->cs;
>>  +       cs->name        = "ingenic-ost";
>>  +       cs->rating      = 320;
>>  +       cs->flags       = CLOCK_SOURCE_IS_CONTINUOUS;
>>  +
>>  +       if (soc_info->is64bit) {
>>  +               cs->mask = CLOCKSOURCE_MASK(64);
>>  +               cs->read = ingenic_ost_clocksource_read;
>>  +       } else {
>>  +               cs->mask = CLOCKSOURCE_MASK(32);
>>  +               cs->read = (u64 (*)(struct clocksource 
>> *))ingenic_ost_read_cnth;
> 
> The function is declared as ingenic_ost_read_cnth(void), are you sure
> this is going to work ?

Hence the cast. The clocksource pointer passed as the first argument 
will
simply be ignored by the function. This works fine.

>>  +       }
>>  +
>>  +       err = clocksource_register_hz(cs, rate);
>>  +       if (err) {
>>  +               dev_err(dev, "clocksource registration failed: 
>> %d\n", err);
>>  +               clk_disable_unprepare(ost->clk);
>>  +               return err;
>>  +       }
>>  +
>>  +       /* Cannot register a sched_clock with interrupts on */
>>  +       local_irq_save(flags);
>>  +       if (soc_info->is64bit)
>>  +               sched_clock_register(ingenic_ost_read_cntl, 32, 
>> rate);
>>  +       else
>>  +               sched_clock_register(ingenic_ost_read_cnth, 32, 
>> rate);
>>  +       local_irq_restore(flags);
>>  +
>>  +       return 0;
>>  +}
>>  +
>>  +#ifdef CONFIG_PM_SLEEP
>>  +static int ingenic_ost_suspend(struct device *dev)
>>  +{
>>  +       struct ingenic_ost *ost = dev_get_drvdata(dev);
>>  +
>>  +       clk_disable(ost->clk);
>>  +       return 0;
>>  +}
>>  +
>>  +static int ingenic_ost_resume(struct device *dev)
>>  +{
>>  +       struct ingenic_ost *ost = dev_get_drvdata(dev);
>>  +
>>  +       return clk_enable(ost->clk);
>>  +}
>>  +
>>  +static SIMPLE_DEV_PM_OPS(ingenic_ost_pm_ops, ingenic_ost_suspend,
>>  +                        ingenic_ost_resume);
>>  +#define INGENIC_OST_PM_OPS (&ingenic_ost_pm_ops)
>>  +#else
>>  +#define INGENIC_OST_PM_OPS NULL
>>  +#endif /* CONFIG_PM_SUSPEND */
>>  +
>>  +static const struct ingenic_ost_soc_info jz4725b_ost_soc_info = {
>>  +       .is64bit = false,
>>  +};
>>  +
>>  +static const struct ingenic_ost_soc_info jz4770_ost_soc_info = {
>>  +       .is64bit = true,
>>  +};
>>  +
>>  +static const struct of_device_id ingenic_ost_of_match[] = {
>>  +       { .compatible = "ingenic,jz4725b-ost", .data = 
>> &jz4725b_ost_soc_info, },
>>  +       { .compatible = "ingenic,jz4770-ost",  .data = 
>> &jz4770_ost_soc_info,  },
>>  +       { }
>>  +};
>>  +
>>  +static struct platform_driver ingenic_ost_driver = {
>>  +       .driver = {
>>  +               .name   = "ingenic-ost",
>>  +               .pm     = INGENIC_OST_PM_OPS,
>>  +               .of_match_table = ingenic_ost_of_match,
>>  +       },
>>  +};
>>  +
>>  +/* FIXME: Using device_initcall (or buildin_platform_driver_probe) 
>> results
>>  + * in the driver not being probed at all. It worked in 4.18...
>>  + */
>>  +static int __init ingenic_ost_drv_register(void)
>>  +{
>>  +       return platform_driver_probe(&ingenic_ost_driver,
>>  +                                    ingenic_ost_probe);
>>  +}
>>  +late_initcall(ingenic_ost_drv_register);
>>  --
>>  2.11.0
>> 


  parent reply	other threads:[~2019-01-23 17:27 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-12 22:08 [PATCH v8 00/26] Ingenic TCU patchset v8 Paul Cercueil
2018-12-12 22:08 ` [PATCH v8 01/26] dt-bindings: ingenic: Add DT bindings for TCU clocks Paul Cercueil
2018-12-17 23:18   ` Stephen Boyd
2018-12-12 22:08 ` [PATCH v8 02/26] doc: Add doc for the Ingenic TCU hardware Paul Cercueil
2018-12-12 22:08 ` [PATCH v8 03/26] dt-bindings: Add doc for the Ingenic TCU drivers Paul Cercueil
2018-12-17 21:05   ` Rob Herring
2018-12-17 22:03     ` Paul Cercueil
2018-12-18 16:36       ` Rob Herring
2018-12-22 11:09         ` Paul Cercueil
2018-12-12 22:08 ` [PATCH v8 04/26] clocksource: Add a new timer-ingenic driver Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 05/26] clocksource: Add driver for the Ingenic JZ47xx OST Paul Cercueil
2019-01-23 12:58   ` Mathieu Malaterre
2019-01-23 14:31     ` Guenter Roeck
2019-01-23 17:25       ` Paul Cercueil
2019-01-23 18:01         ` Guenter Roeck
2019-01-24 19:28           ` Stephen Boyd
2019-01-24 20:46             ` Paul Cercueil
2019-01-24 22:46               ` Stephen Boyd
2019-01-24 22:53                 ` Paul Cercueil
2019-02-23  3:17                   ` Paul Cercueil
2019-02-25 18:05                     ` Stephen Boyd
2019-02-27 23:54                       ` Paul Cercueil
2019-01-23 17:27     ` Paul Cercueil [this message]
2018-12-12 22:09 ` [PATCH v8 06/26] MAINTAINERS: Add myself as maintainer for Ingenic TCU drivers Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 07/26] watchdog: jz4740: Use WDT clock provided by TCU driver Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 08/26] watchdog: jz4740: Use regmap " Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 09/26] watchdog: jz4740: Avoid starting watchdog in set_timeout Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 10/26] watchdog: jz4740: Drop dependency on MACH_JZ47xx, use COMPILE_TEST Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 11/26] pwm: jz4740: Use regmap and clocks from TCU driver Paul Cercueil
2018-12-13  9:30   ` Uwe Kleine-König
2018-12-13 14:34     ` Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 12/26] pwm: jz4740: Allow selection of PWM channels 0 and 1 Paul Cercueil
2018-12-13  9:18   ` Uwe Kleine-König
2018-12-13 13:58     ` Paul Cercueil
2018-12-13 20:32       ` Uwe Kleine-König
2018-12-16 13:36         ` Paul Cercueil
2018-12-17  7:43           ` Uwe Kleine-König
2018-12-12 22:09 ` [PATCH v8 13/26] pwm: jz4740: Drop dependency on MACH_INGENIC, use COMPILE_TEST Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 14/26] pwm: jz4740: Remove unused devicetree compatible strings Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 15/26] pwm: jz4740: Add support for the JZ4725B Paul Cercueil
2018-12-13  9:24   ` Uwe Kleine-König
2018-12-13 14:03     ` Paul Cercueil
2018-12-13 16:18       ` Thierry Reding
2018-12-13 20:42       ` Uwe Kleine-König
2018-12-14 13:50         ` Linus Walleij
2018-12-14 14:26           ` Uwe Kleine-König
2018-12-14 14:56             ` Linus Walleij
2018-12-16 14:18             ` Paul Cercueil
2018-12-17  7:53               ` Uwe Kleine-König
2018-12-20 17:39                 ` Thierry Reding
2018-12-20 20:58                   ` Uwe Kleine-König
2018-12-12 22:09 ` [PATCH v8 16/26] clk: jz4740: Add TCU clock Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 17/26] MIPS: Kconfig: Select TCU timer driver when MACH_INGENIC is set Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 18/26] MIPS: jz4740: Add DTS nodes for the TCU drivers Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 19/26] MIPS: qi_lb60: Move PWM devices to devicetree Paul Cercueil
2018-12-12 22:09 ` [PATCH v8 20/26] MIPS: qi_lb60: Reduce system timer and clocksource to 750 kHz Paul Cercueil
2018-12-12 22:12 ` [PATCH v8 21/26] MIPS: CI20: Reduce system timer and clocksource to 3 MHz Paul Cercueil
2018-12-12 22:12 ` [PATCH v8 22/26] MIPS: CI20: defconfig: enable OST driver Paul Cercueil
2018-12-12 22:13 ` [PATCH v8 23/26] MIPS: GCW0: Move clocksource to TCU channel 2 Paul Cercueil
2018-12-12 22:13 ` [PATCH v8 24/26] MIPS: GCW0: Reduce system timer and clocksource to 750 kHz Paul Cercueil
2018-12-12 22:14 ` [PATCH v8 25/26] MIPS: GCW0: defconfig: Enable OST, watchdog, PWM drivers Paul Cercueil
2018-12-12 22:15 ` [PATCH v8 26/26] MIPS: jz4740: Drop obsolete code Paul Cercueil
2019-01-24 21:26 ` [PATCH v8 00/26] Ingenic TCU patchset v8 Mathieu Malaterre
2019-01-24 21:41   ` Paul Cercueil

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1548264445.3173.2@crapouillou.net \
    --to=paul@crapouillou.net \
    --cc=corbet@lwn.net \
    --cc=daniel.lezcano@linaro.org \
    --cc=devicetree@vger.kernel.org \
    --cc=ezequiel@collabora.co.uk \
    --cc=jhogan@kernel.org \
    --cc=linux-clk@vger.kernel.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@vger.kernel.org \
    --cc=linux-pwm@vger.kernel.org \
    --cc=linux-watchdog@vger.kernel.org \
    --cc=maarten@treewalker.org \
    --cc=malat@debian.org \
    --cc=mark.rutland@arm.com \
    --cc=od@zcrc.me \
    --cc=paul.burton@mips.com \
    --cc=prasannatsmkumar@gmail.com \
    --cc=ralf@linux-mips.org \
    --cc=robh+dt@kernel.org \
    --cc=tglx@linutronix.de \
    --cc=thierry.reding@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).