All of lore.kernel.org
 help / color / mirror / Atom feed
From: baruch@tkos.co.il (Baruch Siach)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v4 2/2] clocksource: driver for Conexant Digicolor SoC timer
Date: Mon, 26 Jan 2015 12:15:11 +0200	[thread overview]
Message-ID: <20150126101511.GC3148@tarshish> (raw)
In-Reply-To: <54C60C4F.50507@linaro.org>

Hi Daniel,

On Mon, Jan 26, 2015 at 10:43:43AM +0100, Daniel Lezcano wrote:
[...]
> >+static struct dc_dev_t {
> >+	void __iomem *timer_base;
> >+	u32 ticks_per_jiffy;
> >+} dc_dev;
> 
> Hi Baruch,
> 
> your code is valid but I think there is a misunderstanding when we talked
> about the encapsulation.
> 
> I was expecting something like:
> 
> static struct digicolor_timer {
> 	void __iomem *base;
> 	u32 ticks_per_jiffy;
> 	int channel;
> 	struct clock_event_device ce;
> };
> 
> struct digicolor_timer *dg_timer(struct clock_event_device *ce)
> {
> 	return container_of(ce, struct digicolor_timer, ce);
> }
> 
> So the timer could be accessed directly from the clock_event_device
> structure. The exception is the 'sched_read' and the init function where you
> will have to address directly the global variable.
> 
> Perhaps, that would be nice if you take the opportunity to encapsulate the
> enable/disable functions.
> 
> static inline void dg_timer_disable(struct clock_event_device *ce)
> {
> 	struct digital_timer *dt = dg_timer(ce);
> 	writeb(CONTROL_DISABLE, dt->base + CONTROL(dt->channel));	
> }
> 
> static inline void dg_timer_enable(struct clock_event_device *ce, int mode)
> {
> 	struct digital_timer *dt = dg_timer(ce);
> 	writeb(CONTROL_ENABLE | mode, dt->base, CONTROL(dt->channel));
> }
> 
> >+
> >+static void digicolor_clkevt_mode(enum clock_event_mode mode,
> >+				  struct clock_event_device *clk)
> 
> IMO, 'clk' argument name is confusing. I would suggest to use 'ce' or
> whatever.
> 
> >+{
> >+	switch (mode) {
> >+	case CLOCK_EVT_MODE_PERIODIC:
> >+		writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> >+		writel(dc_dev.ticks_per_jiffy,
> >+		       dc_dev.timer_base + COUNT(TIMER_C));
> >+		writeb(CONTROL_ENABLE | CONTROL_MODE_PERIODIC,
> >+		       dc_dev.timer_base + CONTROL(TIMER_C));
> 
> Becomes:
> 
> 		dg_timer_disable(ce);
> 		dg_timer_enable(ce, CONTROL_MODE_PERIODIC);
> 
> ...
> 
> >+		break;
> >+	case CLOCK_EVT_MODE_ONESHOT:
> >+		writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> >+		writeb(CONTROL_ENABLE | CONTROL_MODE_ONESHOT,
> >+		       dc_dev.timer_base + CONTROL(TIMER_C));
> 
> 
> Becomes:
> 		dg_timer_disable(ce);
> 		dg_timer_enable(ce, CONTROL_MODE_ONESHOT);
> 	
> 
> >+		break;
> >+	case CLOCK_EVT_MODE_UNUSED:
> >+	case CLOCK_EVT_MODE_SHUTDOWN:
> >+	default:
> >+		writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> 
> Becomes:
> 		dg_timer_disable(ce);
> 
> >+		break;
> >+	}
> >+}
> >+
> >+static int digicolor_clkevt_next_event(unsigned long evt,
> >+				       struct clock_event_device *unused)
> >+{
> >+	writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> >+	writel(evt, dc_dev.timer_base + COUNT(TIMER_C));
> >+	writeb(CONTROL_ENABLE | CONTROL_MODE_ONESHOT,
> >+	       dc_dev.timer_base + CONTROL(TIMER_C));
> 
> Becomes:
> 	dg_timer_disable(ce);
> 	writel(evt, dg_timer(ce)->base + COUNT(dg_timer(ce)->base));
> 	dg_timer_enable(ce, CONTROL_MODE_ONESHOT);
> >+	return 0;
> >+}
> >+
> >+static struct clock_event_device digicolor_clockevent = {
> >+	.name = "digicolor_tick",
> >+	.rating = 340,
> >+	.features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
> >+	.set_mode = digicolor_clkevt_mode,
> >+	.set_next_event = digicolor_clkevt_next_event,
> >+};
> 
> 
> struct digicolor_timer dg_timer = {
> 	
> 	.ce = {
> 		.name = "digicolor_tick",
> 		.rating = 340,
> 		.features = CLOCK_EVT_FEAT_PERIODIC |
> 				CLOCK_EVT_FEAT_ONESHOT,
> 		.set_mode = digicolor_clkevt_mode,
> 		.set_next_event = digicolor_clkevt_next_event,
> 	},
> 	.channel = TIMER_C,
> };
> 
> What do you think ?

OK. I'll give it a try.

Thanks for your prompt response and thorough review.

baruch

-- 
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.2.679.5364, http://www.tkos.co.il -

WARNING: multiple messages have this Message-ID (diff)
From: Baruch Siach <baruch@tkos.co.il>
To: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v4 2/2] clocksource: driver for Conexant Digicolor SoC timer
Date: Mon, 26 Jan 2015 12:15:11 +0200	[thread overview]
Message-ID: <20150126101511.GC3148@tarshish> (raw)
In-Reply-To: <54C60C4F.50507@linaro.org>

Hi Daniel,

On Mon, Jan 26, 2015 at 10:43:43AM +0100, Daniel Lezcano wrote:
[...]
> >+static struct dc_dev_t {
> >+	void __iomem *timer_base;
> >+	u32 ticks_per_jiffy;
> >+} dc_dev;
> 
> Hi Baruch,
> 
> your code is valid but I think there is a misunderstanding when we talked
> about the encapsulation.
> 
> I was expecting something like:
> 
> static struct digicolor_timer {
> 	void __iomem *base;
> 	u32 ticks_per_jiffy;
> 	int channel;
> 	struct clock_event_device ce;
> };
> 
> struct digicolor_timer *dg_timer(struct clock_event_device *ce)
> {
> 	return container_of(ce, struct digicolor_timer, ce);
> }
> 
> So the timer could be accessed directly from the clock_event_device
> structure. The exception is the 'sched_read' and the init function where you
> will have to address directly the global variable.
> 
> Perhaps, that would be nice if you take the opportunity to encapsulate the
> enable/disable functions.
> 
> static inline void dg_timer_disable(struct clock_event_device *ce)
> {
> 	struct digital_timer *dt = dg_timer(ce);
> 	writeb(CONTROL_DISABLE, dt->base + CONTROL(dt->channel));	
> }
> 
> static inline void dg_timer_enable(struct clock_event_device *ce, int mode)
> {
> 	struct digital_timer *dt = dg_timer(ce);
> 	writeb(CONTROL_ENABLE | mode, dt->base, CONTROL(dt->channel));
> }
> 
> >+
> >+static void digicolor_clkevt_mode(enum clock_event_mode mode,
> >+				  struct clock_event_device *clk)
> 
> IMO, 'clk' argument name is confusing. I would suggest to use 'ce' or
> whatever.
> 
> >+{
> >+	switch (mode) {
> >+	case CLOCK_EVT_MODE_PERIODIC:
> >+		writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> >+		writel(dc_dev.ticks_per_jiffy,
> >+		       dc_dev.timer_base + COUNT(TIMER_C));
> >+		writeb(CONTROL_ENABLE | CONTROL_MODE_PERIODIC,
> >+		       dc_dev.timer_base + CONTROL(TIMER_C));
> 
> Becomes:
> 
> 		dg_timer_disable(ce);
> 		dg_timer_enable(ce, CONTROL_MODE_PERIODIC);
> 
> ...
> 
> >+		break;
> >+	case CLOCK_EVT_MODE_ONESHOT:
> >+		writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> >+		writeb(CONTROL_ENABLE | CONTROL_MODE_ONESHOT,
> >+		       dc_dev.timer_base + CONTROL(TIMER_C));
> 
> 
> Becomes:
> 		dg_timer_disable(ce);
> 		dg_timer_enable(ce, CONTROL_MODE_ONESHOT);
> 	
> 
> >+		break;
> >+	case CLOCK_EVT_MODE_UNUSED:
> >+	case CLOCK_EVT_MODE_SHUTDOWN:
> >+	default:
> >+		writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> 
> Becomes:
> 		dg_timer_disable(ce);
> 
> >+		break;
> >+	}
> >+}
> >+
> >+static int digicolor_clkevt_next_event(unsigned long evt,
> >+				       struct clock_event_device *unused)
> >+{
> >+	writeb(CONTROL_DISABLE, dc_dev.timer_base + CONTROL(TIMER_C));
> >+	writel(evt, dc_dev.timer_base + COUNT(TIMER_C));
> >+	writeb(CONTROL_ENABLE | CONTROL_MODE_ONESHOT,
> >+	       dc_dev.timer_base + CONTROL(TIMER_C));
> 
> Becomes:
> 	dg_timer_disable(ce);
> 	writel(evt, dg_timer(ce)->base + COUNT(dg_timer(ce)->base));
> 	dg_timer_enable(ce, CONTROL_MODE_ONESHOT);
> >+	return 0;
> >+}
> >+
> >+static struct clock_event_device digicolor_clockevent = {
> >+	.name = "digicolor_tick",
> >+	.rating = 340,
> >+	.features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
> >+	.set_mode = digicolor_clkevt_mode,
> >+	.set_next_event = digicolor_clkevt_next_event,
> >+};
> 
> 
> struct digicolor_timer dg_timer = {
> 	
> 	.ce = {
> 		.name = "digicolor_tick",
> 		.rating = 340,
> 		.features = CLOCK_EVT_FEAT_PERIODIC |
> 				CLOCK_EVT_FEAT_ONESHOT,
> 		.set_mode = digicolor_clkevt_mode,
> 		.set_next_event = digicolor_clkevt_next_event,
> 	},
> 	.channel = TIMER_C,
> };
> 
> What do you think ?

OK. I'll give it a try.

Thanks for your prompt response and thorough review.

baruch

-- 
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch@tkos.co.il - tel: +972.2.679.5364, http://www.tkos.co.il -

  reply	other threads:[~2015-01-26 10:15 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-26  7:56 [PATCH v4 0/2] clocksource: Conexant CX92755 timers support Baruch Siach
2015-01-26  7:56 ` Baruch Siach
2015-01-26  7:56 ` [PATCH v4 1/2] clocksource: devicetree: document Conexant Digicolor timer binding Baruch Siach
2015-01-26  7:56   ` Baruch Siach
2015-01-26  7:56 ` [PATCH v4 2/2] clocksource: driver for Conexant Digicolor SoC timer Baruch Siach
2015-01-26  7:56   ` Baruch Siach
2015-01-26  9:43   ` Daniel Lezcano
2015-01-26  9:43     ` Daniel Lezcano
2015-01-26 10:15     ` Baruch Siach [this message]
2015-01-26 10:15       ` Baruch Siach
2015-01-26 10:26       ` Daniel Lezcano
2015-01-26 10:26         ` Daniel Lezcano
2015-01-26 10:38       ` Thomas Gleixner
2015-01-26 10:38         ` Thomas Gleixner

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=20150126101511.GC3148@tarshish \
    --to=baruch@tkos.co.il \
    --cc=linux-arm-kernel@lists.infradead.org \
    /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 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.