All of lore.kernel.org
 help / color / mirror / Atom feed
From: wagi@monom.org (Daniel Wagner)
To: cip-dev@lists.cip-project.org
Subject: [cip-dev] iWave RZ/G1M hrtimers
Date: Fri, 4 May 2018 12:25:47 +0200	[thread overview]
Message-ID: <45738fc6-fe53-8309-77ab-be6204fdc68c@monom.org> (raw)
In-Reply-To: <288bed25-0560-69a2-7239-d25387f9ab64@monom.org>

On 05/04/2018 12:16 PM, Daniel Wagner wrote:
> Hi Biju,
> 
> Not sure if I ping the right person about this problem.
> 
> I am doing some realtime testing on the iWave RZ/G1M board and noticed
> quite high latency values for cyclictest (around 3ms).
> 
> As it turns out there is no hrtimer clock source available. Just the
> arch cp8 clock source. The timer subsystem then decides to set the
> resolution of the clocksource to 4000000 nsecs which explains the rather
> bad value for cyclictest.
> 
> 
> clocksource: arch_sys_counter: mask: 0xffffffffffffff max_cycles:
> 0x24e6a1710, max_idle_ns: 440795202120 ns
> sched_clock: 56 bits at 10MHz, resolution 100ns, wraps every 4398046511100ns
> Switching to timer-based delay loop, resolution 100ns
> 
> # cat current_clocksource
> arch_sys_counter
> 
> # cat available_clocksource
> arch_sys_counter jiffies
> 
> # cat /proc/timer_list
> Timer List Version: v0.8
> HRTIMER_MAX_CLOCK_BASES: 4
> now at 1705568146462 nsecs
> 
> cpu: 0
>    clock 0:
>     .base:       dfbcbe40
>     .index:      0
>     .resolution: 4000000 nsecs
>     .get_time:   ktime_get
>     .offset:     0 nsecs
> 
> 
> The question is do I miss some magic CONFIG option or is the support not
> yet in v4.4-cip ?

Forgot to mention. I am using the shmobile_defconfig configuration with 
some mods on top of it:

# systemd stuff
CONFIG_CGROUPS=y
CONFIG_CGROUP_DEBUG=n
CONFIG_CGROUP_FREEZER=n
CONFIG_INOTIFY_USER=y
CONFIG_FHANDLE=y
CONFIG_CRYPTO_USER_API_HASH=y
CONFIG_CRYPTO_HMAC=y
CONFIG_AUTOFS4_FS=y
CONFIG_TMPFS_XATTR=y

# -rt tunnings
CONFIG_DEBUG_ATOMIC_SLEEP=y
CONFIG_DEBUG_LOCK_ALLOC=n
CONFIG_PROVE_LOCKING=n
CONFIG_LOCKDEP=n

CONFIG_CPU_FREQ=n
CONFIG_CPU_IDLE=n

CONFIG_NO_HZ=n
CONFIG_HZ_PERIODIC=y

CONFIG_HZ_250=y
CONFIG_HZ=250

CONFIG_HIGH_RES_TIMERS=y

CONFIG_SUSPEND=n
CONFIG_HIBERNATION=n
CONFIG_PM=n

CONFIG_PREEMPT_RT_FULL=y

# root nfs
CONFIG_NFS_FS=y
CONFIG_NFS_V4=y
CONFIG_NFS_V4_1=y
CONFIG_NFS_V4_2=y

# some debugging stuff
CONFIG_FTRACE=y
CONFIG_FUNCTION_TRACER=y
CONFIG_SCHED_TRACER=y

  reply	other threads:[~2018-05-04 10:25 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-04 10:16 [cip-dev] iWave RZ/G1M hrtimers Daniel Wagner
2018-05-04 10:25 ` Daniel Wagner [this message]
2018-05-04 11:22   ` Biju Das
2018-05-04 11:29     ` Daniel Wagner

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=45738fc6-fe53-8309-77ab-be6204fdc68c@monom.org \
    --to=wagi@monom.org \
    --cc=cip-dev@lists.cip-project.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.