[RFC][PATCH 0/3] Last of the clocksource_register_hz cleanups

[RFC][PATCH 0/3] Last of the clocksource_register_hz cleanups

[John Stultz]

So I'm hoping to finally kill off clocksource_register(), so I wanted
to get the last of the hardware specific clocksources changes that I
cannot test to be reviewed by maintainers.

In some cases the patches are not clearly correct, or the existing
code is broken and I don't know how to fix it. So feedback here
would be greatly appreciated!

thanks
-john

CC: Martin Schwidefsky <schwidefsky@de.ibm.com>
CC: Jim Cromie <jim.cromie@gmail.com>
CC: Ralf Baechle <ralf@linux-mips.org>
CC: Yong Zhang <yong.zhang0@gmail.com>

John Stultz (3):
  clocksource: s390: Convert s390 to use clocksource_register_khz
  clocksource: scx200_hrt: Convert scx200 to use
    clocksource_register_hz
  clocksource: Convert mips pnx8550 to use clocksource_register_hz

 arch/mips/pnx8550/common/time.c  |    7 ++++++-
 arch/s390/kernel/time.c          |   13 ++++++++++---
 drivers/clocksource/scx200_hrt.c |   19 ++++++-------------
 3 files changed, 22 insertions(+), 17 deletions(-)

-- 
1.7.3.2.146.gca209

[RFC][PATCH 1/3] clocksource: s390: Convert s390 to use clocksource_register_khz

[John Stultz]

Converts s390 TOD clock to use clocksource_register_khz

The tod freq I calculated seems a bit fast. Need someone
to double check me here.

CC: Martin Schwidefsky <schwidefsky@de.ibm.com>
CC: linux-s390@vger.kernel.org
Signed-off-by: John Stultz <john.stultz@linaro.org>
---
 arch/s390/kernel/time.c |   13 ++++++++++---
 1 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index fa02f44..37e8ec5 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -207,8 +207,6 @@ static struct clocksource clocksource_tod = {
 	.rating		= 400,
 	.read		= read_tod_clock,
 	.mask		= -1ULL,
-	.mult		= 1000,
-	.shift		= 12,
 	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
 };
 
@@ -267,7 +265,16 @@ void __init time_init(void)
 	if (register_external_interrupt(0x1406, timing_alert_interrupt))
 		panic("Couldn't request external interrupt 0x1406");
 
-	if (clocksource_register(&clocksource_tod) != 0)
+	/* s390 TOD apparently runs at 4,096,000,000 cycs/sec
+	 * XXX - Brute forced tod freq from:
+	 *    2^shift/mult * NSEC_PER_SEC	= cyc/sec
+	 *    2^12 / 1000  * 1000000000		= cyc/sec
+	 *    2^12 * 1000000			= cyc/sec
+	 *    2^12 * 1000			= kcyc/sec
+	 *    1000<<12				= kcyc/sec
+	 * Could be totally wrong. -johnstul
+	 */
+	if (clocksource_register_khz(&clocksource_tod, (1000<<12) ) != 0)
 		panic("Could not register TOD clock source");
 
 	/* Enable TOD clock interrupts on the boot cpu. */
-- 
1.7.3.2.146.gca209

[RFC][PATCH 2/3] clocksource: scx200_hrt: Convert scx200 to use clocksource_register_hz

[John Stultz]

Converts the scx200 clocksource to using clocksource_register_hz.

This wasn't a straight conversion, so would like some extra review
by the maintainers.

CC: Jim Cromie <jim.cromie@gmail.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
---
 drivers/clocksource/scx200_hrt.c |   19 ++++++-------------
 1 files changed, 6 insertions(+), 13 deletions(-)

diff --git a/drivers/clocksource/scx200_hrt.c b/drivers/clocksource/scx200_hrt.c
index 27f4d96..24f884c 100644
--- a/drivers/clocksource/scx200_hrt.c
+++ b/drivers/clocksource/scx200_hrt.c
@@ -49,9 +49,6 @@ static cycle_t read_hrt(struct clocksource *cs)
 	return (cycle_t) inl(scx200_cb_base + SCx200_TIMER_OFFSET);
 }
 
-#define HRT_SHIFT_1	22
-#define HRT_SHIFT_27	26
-
 static struct clocksource cs_hrt = {
 	.name		= "scx200_hrt",
 	.rating		= 250,
@@ -63,6 +60,7 @@ static struct clocksource cs_hrt = {
 
 static int __init init_hrt_clocksource(void)
 {
+	u32 freq;
 	/* Make sure scx200 has initialized the configuration block */
 	if (!scx200_cb_present())
 		return -ENODEV;
@@ -79,19 +77,14 @@ static int __init init_hrt_clocksource(void)
 	outb(HR_TMEN | (mhz27 ? HR_TMCLKSEL : 0),
 	     scx200_cb_base + SCx200_TMCNFG_OFFSET);
 
-	if (mhz27) {
-		cs_hrt.shift = HRT_SHIFT_27;
-		cs_hrt.mult = clocksource_hz2mult((HRT_FREQ + ppm) * 27,
-						  cs_hrt.shift);
-	} else {
-		cs_hrt.shift = HRT_SHIFT_1;
-		cs_hrt.mult = clocksource_hz2mult(HRT_FREQ + ppm,
-						  cs_hrt.shift);
-	}
+	freq = (HRT_FREQ + ppm);
+	if (mhz27)
+		freq *= 27;
+
 	printk(KERN_INFO "enabling scx200 high-res timer (%s MHz +%d ppm)\n",
 		mhz27 ? "27":"1", ppm);
 
-	return clocksource_register(&cs_hrt);
+	return clocksource_register_hz(&cs_hrt, freq);
 }
 
 module_init(init_hrt_clocksource);
-- 
1.7.3.2.146.gca209

[PATCH 3/3] clocksource: Convert mips pnx8550 to use clocksource_register_hz

[John Stultz]

This patch converts the registration to use clocksource_register_hz()
but is clearly broken, as the pnx_clocksource doesn't seem to set a
proper mult/shift pair.

CC: Ralf Baechle <ralf@linux-mips.org>
CC: Yong Zhang <yong.zhang0@gmail.com>
CC: linux-mips@linux-mips.org
Signed-off-by: John Stultz <john.stultz@linaro.org>
---
 arch/mips/pnx8550/common/time.c |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletions(-)

diff --git a/arch/mips/pnx8550/common/time.c b/arch/mips/pnx8550/common/time.c
index 831d6b3..006ac88 100644
--- a/arch/mips/pnx8550/common/time.c
+++ b/arch/mips/pnx8550/common/time.c
@@ -104,7 +104,12 @@ __init void plat_time_init(void)
 
 	pnx8xxx_clockevent.cpumask = cpu_none_mask;
 	clockevents_register_device(&pnx8xxx_clockevent);
-	clocksource_register(&pnx_clocksource);
+	
+	/*
+	 * XXX - Nothing seems to set pnx_clocksource mult/shift! 
+	 * So I don't know what freq to use here. Help! -johnstul
+	 */
+	clocksource_register_hz(&pnx_clocksource, 0);
 
 	/* Timer 1 start */
 	configPR = read_c0_config7();
-- 
1.7.3.2.146.gca209

Re: [RFC][PATCH 2/3] clocksource: scx200_hrt: Convert scx200 to use clocksource_register_hz

[Jim Cromie]

[-- Attachment #1: Type: text/plain, Size: 5918 bytes --]

On Mon, Jan 23, 2012 at 5:05 PM, John Stultz <john.stultz@linaro.org> wrote:
> Converts the scx200 clocksource to using clocksource_register_hz.
>
> This wasn't a straight conversion, so would like some extra review
> by the maintainers.
>

hi John,

very nice.  Tested with mhz27 too.

Signed-off-by: Jim Cromie <jim.cromie@gmail.com>

It eliminates this noise that Ive started seeing recently (for some
value of recent)

enabling scx200 high-res timer (1 MHz +0 ppm)
------------[ cut here ]------------
WARNING: at /home/jimc/projects/lx/linux-2.6/kernel/time/clocksource.c:740
clocksource_register+0x78/0xbd()
Clocksource scx200_hrt might overflow on 11% adjustment
Modules linked in: scx200_hrt(+) scx200 pc8736x_gpio nsc_gpio pc87360
hwmon_vid scx200_acb ohci_hcd pata_sc1200
Pid: 291, comm: modprobe Not tainted 3.2.0-skd-00063-g296c136 #256
Call Trace:
 [<c01201a1>] warn_slowpath_common+0x7c/0x91
 [<c0140266>] ? clocksource_register+0x78/0xbd
 [<c0140266>] ? clocksource_register+0x78/0xbd
 [<c0120241>] warn_slowpath_fmt+0x33/0x35
 [<c0140266>] clocksource_register+0x78/0xbd
 [<c886412d>] init_hrt_clocksource+0x12d/0x139 [scx200_hrt]
 [<c010108b>] do_one_initcall+0x8b/0x138
 [<c8864000>] ? 0xc8863fff
 [<c014a6db>] sys_init_module+0x13ef/0x160e
 [<c042015c>] syscall_call+0x7/0xb
---[ end trace 69c8ca76cf89ab2d ]---
Switching to clocksource scx200_hrt
Loading kernel modules...chrdev_alloc_region: chrdev_alloc_region
reserves a dynamic major, but was called with major:3077.  Caller
should pass major:0 for this.
done.
------------[ cut here ]------------
WARNING: at /home/jimc/projects/lx/linux-2.6/kernel/time/timekeeping.c:863
do_timer+0x56f/0x6ae()
Adjusting scx200_hrt more then 11% (-100663808 vs 360710144)

NB - s/then/than/   Im attaching a patch for that.
Also, "when then" is a little awkward.
		/*
		 * Have to be careful here. If we're in oneshot mode,
		 * before we call tick_periodic() in a loop, we need
		 * to be sure we're using a real hardware clocksource.
		 * Otherwise we could get trapped in an infinite
		 * loop, as the tick_periodic() increments jiffies,
		 * when then will increment time, posibly causing
		 * the loop to trigger again and again.
		 */


Modules linked in: scx200_gpio scx200_hrt scx200 pc8736x_gpio nsc_gpio
pc87360 hwmon_vid scx200_acb ohci_hcd pata_sc1200
Pid: 58, comm: collector Tainted: G        W    3.2.0-skd-00063-g296c136 #256
Call Trace:
 [<c01201a1>] warn_slowpath_common+0x7c/0x91
 [<c013ebbe>] ? do_timer+0x56f/0x6ae
 [<c013ebbe>] ? do_timer+0x56f/0x6ae
 [<c0120241>] warn_slowpath_fmt+0x33/0x35
 [<c013ebbe>] do_timer+0x56f/0x6ae
 [<c014319b>] tick_do_update_jiffies64+0xc0/0xee
 [<c01432e1>] tick_sched_timer+0x45/0x9e
 [<c0139c01>] ? __remove_hrtimer+0x2a/0x87
 [<c0139d81>] __run_hrtimer+0x93/0x131
 [<c014329c>] ? tick_nohz_handler+0xd3/0xd3
 [<c013a549>] hrtimer_interrupt+0xe1/0x1f4
 [<c010346d>] timer_interrupt+0x18/0x1f
 [<c015fbcf>] handle_irq_event_percpu+0x4a/0x197
 [<c014bcfe>] ? pm_qos_request+0xe/0x16
 [<c0104155>] ? mask_and_ack_8259A+0x77/0xde
 [<c0161a3c>] ? unmask_irq+0x23/0x23
 [<c015fd58>] handle_irq_event+0x3c/0x54
 [<c0161a8e>] handle_level_irq+0x52/0x80
 <IRQ>  [<c0102f2c>] ? do_IRQ+0x39/0x87
 [<c02b561a>] ? disk_seqf_start+0x2e/0x84
 [<c0424c70>] ? common_interrupt+0x30/0x40
 [<c01c007b>] ? do_vfs_ioctl+0x30c/0x481
 [<c01f21e9>] ? proc_readdir_de+0x166/0x186
 [<c01c0324>] ? filldir64+0xcc/0xcc
 [<c01f222a>] ? proc_readdir+0x21/0x25
 [<c01c0324>] ? filldir64+0xcc/0xcc
 [<c01edbe9>] ? proc_root_readdir+0x1f/0x41
 [<c01c0509>] ? vfs_readdir+0x6a/0x8f
 [<c01c0324>] ? filldir64+0xcc/0xcc
 [<c01c05e7>] ? sys_getdents+0x63/0x9f
 [<c042015c>] ? syscall_call+0x7/0xb
 [<c0420000>] ? rwsem_down_failed_common+0x5f/0xbc
---[ end trace 69c8ca76cf89ab2e ]---




> CC: Jim Cromie <jim.cromie@gmail.com>
> Signed-off-by: John Stultz <john.stultz@linaro.org>
> ---
>  drivers/clocksource/scx200_hrt.c |   19 ++++++-------------
>  1 files changed, 6 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/clocksource/scx200_hrt.c b/drivers/clocksource/scx200_hrt.c
> index 27f4d96..24f884c 100644
> --- a/drivers/clocksource/scx200_hrt.c
> +++ b/drivers/clocksource/scx200_hrt.c
> @@ -49,9 +49,6 @@ static cycle_t read_hrt(struct clocksource *cs)
>        return (cycle_t) inl(scx200_cb_base + SCx200_TIMER_OFFSET);
>  }
>
> -#define HRT_SHIFT_1    22
> -#define HRT_SHIFT_27   26
> -
>  static struct clocksource cs_hrt = {
>        .name           = "scx200_hrt",
>        .rating         = 250,
> @@ -63,6 +60,7 @@ static struct clocksource cs_hrt = {
>
>  static int __init init_hrt_clocksource(void)
>  {
> +       u32 freq;
>        /* Make sure scx200 has initialized the configuration block */
>        if (!scx200_cb_present())
>                return -ENODEV;
> @@ -79,19 +77,14 @@ static int __init init_hrt_clocksource(void)
>        outb(HR_TMEN | (mhz27 ? HR_TMCLKSEL : 0),
>             scx200_cb_base + SCx200_TMCNFG_OFFSET);
>
> -       if (mhz27) {
> -               cs_hrt.shift = HRT_SHIFT_27;
> -               cs_hrt.mult = clocksource_hz2mult((HRT_FREQ + ppm) * 27,
> -                                                 cs_hrt.shift);
> -       } else {
> -               cs_hrt.shift = HRT_SHIFT_1;
> -               cs_hrt.mult = clocksource_hz2mult(HRT_FREQ + ppm,
> -                                                 cs_hrt.shift);
> -       }
> +       freq = (HRT_FREQ + ppm);
> +       if (mhz27)
> +               freq *= 27;
> +
>        printk(KERN_INFO "enabling scx200 high-res timer (%s MHz +%d ppm)\n",
>                mhz27 ? "27":"1", ppm);
>
> -       return clocksource_register(&cs_hrt);
> +       return clocksource_register_hz(&cs_hrt, freq);
>  }
>
>  module_init(init_hrt_clocksource);
> --
> 1.7.3.2.146.gca209
>

[-- Attachment #2: 0001-time-s-then-than-where-appropriate.patch --]
[-- Type: text/x-patch, Size: 4037 bytes --]

From f75b23207d5729a9ec72ba0d05d9a07e86291b1b Mon Sep 17 00:00:00 2001
From: Jim Cromie <jim.cromie@gmail.com>
Date: Tue, 24 Jan 2012 11:47:55 -0700
Subject: [PATCH] time: s/then/than/ where appropriate


Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
---
 kernel/time/clocksource.c |    2 +-
 kernel/time/timekeeping.c |   16 ++++++++--------
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index d3ad022..ebf98fb 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -500,7 +500,7 @@ static u32 clocksource_max_adjustment(struct clocksource *cs)
 {
 	u64 ret;
 	/*
-	 * We won't try to correct for more then 11% adjustments (110,000 ppm),
+	 * We won't try to correct for more than 11% adjustments (110,000 ppm),
 	 */
 	ret = (u64)cs->mult * 11;
 	do_div(ret,100);
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 2378413..e77c168 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -807,7 +807,7 @@ static void timekeeping_adjust(s64 offset)
 	int adj;
 
 	/*
-	 * The point of this is to check if the error is greater then half
+	 * The point of this is to check if the error is greater than half
 	 * an interval.
 	 *
 	 * First we shift it down from NTP_SHIFT to clocksource->shifted nsecs.
@@ -815,7 +815,7 @@ static void timekeeping_adjust(s64 offset)
 	 * Note we subtract one in the shift, so that error is really error*2.
 	 * This "saves" dividing(shifting) intererval twice, but keeps the
 	 * (error > interval) comparision as still measuring if error is
-	 * larger then half an interval.
+	 * larger than half an interval.
 	 *
 	 * Note: It does not "save" on aggrivation when reading the code.
 	 */
@@ -823,7 +823,7 @@ static void timekeeping_adjust(s64 offset)
 	if (error > interval) {
 		/*
 		 * We now divide error by 4(via shift), which checks if
-		 * the error is greater then twice the interval.
+		 * the error is greater than twice the interval.
 		 * If it is greater, we need a bigadjust, if its smaller,
 		 * we can adjust by 1.
 		 */
@@ -857,7 +857,7 @@ static void timekeeping_adjust(s64 offset)
 	WARN_ONCE(timekeeper.clock->maxadj &&
 			(timekeeper.mult + adj > timekeeper.clock->mult +
 						timekeeper.clock->maxadj),
-			"Adjusting %s more then 11%% (%ld vs %ld)\n",
+			"Adjusting %s more than 11%% (%ld vs %ld)\n",
 			timekeeper.clock->name, (long)timekeeper.mult + adj,
 			(long)timekeeper.clock->mult +
 				timekeeper.clock->maxadj);
@@ -932,7 +932,7 @@ static cycle_t logarithmic_accumulation(cycle_t offset, int shift)
 	u64 nsecps = (u64)NSEC_PER_SEC << timekeeper.shift;
 	u64 raw_nsecs;
 
-	/* If the offset is smaller then a shifted interval, do nothing */
+	/* If the offset is smaller than a shifted interval, do nothing */
 	if (offset < timekeeper.cycle_interval<<shift)
 		return offset;
 
@@ -995,13 +995,13 @@ static void update_wall_time(void)
 	 * With NO_HZ we may have to accumulate many cycle_intervals
 	 * (think "ticks") worth of time at once. To do this efficiently,
 	 * we calculate the largest doubling multiple of cycle_intervals
-	 * that is smaller then the offset. We then accumulate that
+	 * that is smaller than the offset. We then accumulate that
 	 * chunk in one go, and then try to consume the next smaller
 	 * doubled multiple.
 	 */
 	shift = ilog2(offset) - ilog2(timekeeper.cycle_interval);
 	shift = max(0, shift);
-	/* Bound shift to one less then what overflows tick_length */
+	/* Bound shift to one less than what overflows tick_length */
 	maxshift = (8*sizeof(tick_length) - (ilog2(tick_length)+1)) - 1;
 	shift = min(shift, maxshift);
 	while (offset >= timekeeper.cycle_interval) {
@@ -1047,7 +1047,7 @@ static void update_wall_time(void)
 
 	/*
 	 * Finally, make sure that after the rounding
-	 * xtime.tv_nsec isn't larger then NSEC_PER_SEC
+	 * xtime.tv_nsec isn't larger than NSEC_PER_SEC
 	 */
 	if (unlikely(xtime.tv_nsec >= NSEC_PER_SEC)) {
 		xtime.tv_nsec -= NSEC_PER_SEC;
-- 
1.7.8.1

Re: [RFC][PATCH 2/3] clocksource: scx200_hrt: Convert scx200 to use clocksource_register_hz

[John Stultz]

On Tue, 2012-01-24 at 11:53 -0700, Jim Cromie wrote:
> On Mon, Jan 23, 2012 at 5:05 PM, John Stultz <john.stultz@linaro.org> wrote:
> > Converts the scx200 clocksource to using clocksource_register_hz.
> >
> > This wasn't a straight conversion, so would like some extra review
> > by the maintainers.
> >
> 
> hi John,
> 
> very nice.  Tested with mhz27 too.
> 
> Signed-off-by: Jim Cromie <jim.cromie@gmail.com>

I assume you mean Acked-by:/Tested-by: ?  Or are you going to push this
upstream?

thanks
-john

Re: [RFC][PATCH 2/3] clocksource: scx200_hrt: Convert scx200 to use clocksource_register_hz

[Jim Cromie]

On Tue, Jan 24, 2012 at 4:00 PM, John Stultz <john.stultz@linaro.org> wrote:
> On Tue, 2012-01-24 at 11:53 -0700, Jim Cromie wrote:
>> On Mon, Jan 23, 2012 at 5:05 PM, John Stultz <john.stultz@linaro.org> wrote:
>> > Converts the scx200 clocksource to using clocksource_register_hz.
>> >
>> > This wasn't a straight conversion, so would like some extra review
>> > by the maintainers.
>> >
>>
>> hi John,
>>
>> very nice.  Tested with mhz27 too.
>>
>> Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
>
> I assume you mean Acked-by:/Tested-by: ?  Or are you going to push this
> upstream?
>

huh - I thought SOB was appropriate since Im maintainer of scx200_hrt.
That said, youre maintainer of TIMEKEEPING, and author of the patch,
so I consider you upstream.  Lets make it instead:

Acked-by: Jim Cromie <jim.cromie@gmail.com>
Tested-by: Jim Cromie <jim.cromie@gmail.com>

Can you also push that trivial s/then/than/ patch I attached,
and/or should I resend it inline, with git-send-email ?

> thanks
> -john
>
>

Re: [RFC][PATCH 2/3] clocksource: scx200_hrt: Convert scx200 to use clocksource_register_hz

[John Stultz]

On Tue, 2012-01-24 at 16:51 -0700, Jim Cromie wrote:
> On Tue, Jan 24, 2012 at 4:00 PM, John Stultz <john.stultz@linaro.org> wrote:
> > On Tue, 2012-01-24 at 11:53 -0700, Jim Cromie wrote:
> >> very nice.  Tested with mhz27 too.
> >>
> >> Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
> >
> > I assume you mean Acked-by:/Tested-by: ?  Or are you going to push this
> > upstream?
> >
> 
> huh - I thought SOB was appropriate since Im maintainer of scx200_hrt.
> That said, youre maintainer of TIMEKEEPING, and author of the patch,
> so I consider you upstream.  Lets make it instead:
> 
> Acked-by: Jim Cromie <jim.cromie@gmail.com>
> Tested-by: Jim Cromie <jim.cromie@gmail.com>

Thanks for clarifying. It might be good to review section 12-14 here (I
have to re-read it periodically myself):
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob_plain;f=Documentation/SubmittingPatches

If you're planning on distributing the patch yourself (pushing it up
towards Linus or giving it to someone else via a git tree or something),
then adding the signed-off-by would be appropriate (although not
necessarily in an email reply - usually just better to tell folks you've
queued it) before sending it onward.

Otherwise if you're not in the distribution path, an Acked-by: would be
more appropriate.

> Can you also push that trivial s/then/than/ patch I attached,
> and/or should I resend it inline, with git-send-email ?

Yep. I've just queued it into my tree. Thanks for that!

thanks
-john