linux-mips.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Wu Zhangjin <wuzhangjin@gmail.com>
To: Ralf Baechle <ralf@linux-mips.org>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>,
	David Daney <david.s.daney@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	linux-mips@linux-mips.org
Subject: Re: [PATCH v6] MIPS: Add a high resolution sched_clock() via cnt32_to_63().
Date: Mon, 18 Jan 2010 15:54:44 +0800	[thread overview]
Message-ID: <1263801284.11671.50.camel@falcon> (raw)
In-Reply-To: <1259319110-16107-1-git-send-email-wuzhangjin@gmail.com>

If the processor support dynamic cpu frequency and the support is
enabled in kernel, this sched_clock() implementation will be broken(and
If the frequency of the MIPS CP0 counter is related to the cpu's
frequency).

So, some extra resitrictions should be added to it.

arch/mips/Kconfig

config CPU_HAS_FIXED_CP0_COUNTER
	bool

config SYS_SUPPORTS_HRES_SCHED_CLOCK
	bool
	depends on CPU_HAS_FIXED_CP0_COUNTER || !CPU_FREQ

arch/mips/kernel/csrc-r4k.c

#ifdef SYS_SUPPORTS_HRES_SCHED_CLOCK

/* The high resolution version of sched_clock() */

#endif

And I'm not sure whether the cavium octeon support dynamic cpu
frequency, if yes, it's high resolution version of sched_clock() also
should be wrapped with the above macro to ensure it is not broken:

arch/mips/cavium-octeon/csrc-octeon.c

Regards,
	Wu Zhangjin

On Fri, 2009-11-27 at 18:51 +0800, Wu Zhangjin wrote:
> From: Wu Zhangjin <wuzhangjin@gmail.com>
> 
> (The changes of this v6 revision from v5 revision:
> 
>  o hard-codes the cycle2ns_scale_factor as 8 for 30(cs->shift) is too
>  big. With 30, the return value of sched_clock() will also overflow quickly.
>  o moves the sched_clock() back into csrc-r4k.c as David and Sergei
>  recommended.
>  o inits c0 count as zero for PRINTK_TIME=y.
>  o drops the HR_SCHED_CLCOK option for the current sched_clock() is stable
>  enough to replace the jiffies based one.
> )
> 
> This patch adds a cnt32_to_63() and MIPS c0 count based sched_clock(),
> which provides high resolution.
> 
> Without it, the Ftrace for MIPS will give useless timestamp information.
> 
> Because cnt32_to_63() needs to be called at least once per half period
> to work properly, Differ from the old version, this v2 revision set up a
> kernel timer to ensure the requirement of some MIPSs which have short c0
> count period.
> 
> And also, we init the c0 count as ZERO(just as jiffies does) in
> time_init() before plat_time_init(), without it, PRINTK_TIME=y will get
> wrong timestamp information. (NOTE: some platforms have initiazlied c0
> count as zero, but some not, this may introduce some duplication,
> perhaps a new patch is needed to remove the initialized of c0 count in
> the platforms later?)
> 
> This is originally from arch/arm/plat-orion/time.c
> 
> This revision works well for function graph tracer now, and also,
> PRINTK_TIME=y will get normal timestamp informatin.
> 
> Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
> ---
>  arch/mips/kernel/csrc-r4k.c |   54 +++++++++++++++++++++++++++++++++++++++++++
>  arch/mips/kernel/time.c     |    3 ++
>  2 files changed, 57 insertions(+), 0 deletions(-)
> 
> diff --git a/arch/mips/kernel/csrc-r4k.c b/arch/mips/kernel/csrc-r4k.c
> index e95a3cd..12755f2 100644
> --- a/arch/mips/kernel/csrc-r4k.c
> +++ b/arch/mips/kernel/csrc-r4k.c
> @@ -6,10 +6,62 @@
>   * Copyright (C) 2007 by Ralf Baechle
>   */
>  #include <linux/clocksource.h>
> +#include <linux/cnt32_to_63.h>
>  #include <linux/init.h>
> +#include <linux/timer.h>
>  
>  #include <asm/time.h>
>  
> +/*
> + * MIPS sched_clock implementation.
> + *
> + * Because the hardware timer period is quite short and because cnt32_to_63()
> + * needs to be called at least once per half period to work properly, a kernel
> + * timer is set up to ensure this requirement is always met.
> + *
> + * Please refer to include/linux/cnt32_to_63.h and arch/arm/plat-orion/time.c
> + */
> +#define CLOCK2NS_SCALE_FACTOR 8
> +
> +static unsigned long clock2ns_scale;
> +
> +unsigned long long notrace sched_clock(void)
> +{
> +	unsigned long long v = cnt32_to_63(read_c0_count());
> +	return (v * clock2ns_scale) >> CLOCK2NS_SCALE_FACTOR;
> +}
> +
> +static struct timer_list cnt32_to_63_keepwarm_timer;
> +
> +static void cnt32_to_63_keepwarm(unsigned long data)
> +{
> +	mod_timer(&cnt32_to_63_keepwarm_timer, round_jiffies(jiffies + data));
> +	sched_clock();
> +}
> +
> +static void __init setup_hres_sched_clock(unsigned long clock)
> +{
> +	unsigned long long v;
> +	unsigned long data;
> +
> +	v = NSEC_PER_SEC;
> +	v <<= CLOCK2NS_SCALE_FACTOR;
> +	v += clock/2;
> +	do_div(v, clock);
> +	/*
> +	 * We want an even value to automatically clear the top bit
> +	 * returned by cnt32_to_63() without an additional run time
> +	 * instruction. So if the LSB is 1 then round it up.
> +	 */
> +	if (v & 1)
> +		v++;
> +	clock2ns_scale = v;
> +
> +	data = 0x80000000UL / clock * HZ;
> +	setup_timer(&cnt32_to_63_keepwarm_timer, cnt32_to_63_keepwarm, data);
> +	mod_timer(&cnt32_to_63_keepwarm_timer, round_jiffies(jiffies + data));
> +}
> +
>  static cycle_t c0_hpt_read(struct clocksource *cs)
>  {
>  	return read_c0_count();
> @@ -27,6 +79,8 @@ int __init init_r4k_clocksource(void)
>  	if (!cpu_has_counter || !mips_hpt_frequency)
>  		return -ENXIO;
>  
> +	setup_hres_sched_clock(mips_hpt_frequency);
> +
>  	/* Calculate a somewhat reasonable rating value */
>  	clocksource_mips.rating = 200 + mips_hpt_frequency / 10000000;
>  
> diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c
> index 1f467d5..4b5e93c 100644
> --- a/arch/mips/kernel/time.c
> +++ b/arch/mips/kernel/time.c
> @@ -152,6 +152,9 @@ static __init int cpu_has_mfc0_count_bug(void)
>  
>  void __init time_init(void)
>  {
> +	if (!mips_clockevent_init() || !cpu_has_mfc0_count_bug())
> +		write_c0_count(0);
> +
>  	plat_time_init();
>  
>  	if (!mips_clockevent_init() || !cpu_has_mfc0_count_bug())

  reply	other threads:[~2010-01-18  7:55 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-11-27 10:51 [PATCH v6] MIPS: Add a high resolution sched_clock() via cnt32_to_63() Wu Zhangjin
2010-01-18  7:54 ` Wu Zhangjin [this message]
2010-01-18 18:08   ` David Daney
2010-01-19  6:42     ` Wu Zhangjin

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=1263801284.11671.50.camel@falcon \
    --to=wuzhangjin@gmail.com \
    --cc=david.s.daney@gmail.com \
    --cc=linux-mips@linux-mips.org \
    --cc=ralf@linux-mips.org \
    --cc=sshtylyov@ru.mvista.com \
    --cc=tglx@linutronix.de \
    /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).