gpsd-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gpsd-dev] Fwd: Re: gpsd for time sync on Ubuntu remix too hard for


From: Gary E. Miller
Subject: Re: [gpsd-dev] Fwd: Re: gpsd for time sync on Ubuntu remix too hard for normals - how to improve?
Date: Wed, 8 May 2019 10:16:28 -0700

Yo Lisandro!

On Wed, 08 May 2019 11:33:02 -0300
Lisandro Damián Nicanor Pérez Meyer <address@hidden> wrote:

> El miércoles, 8 de mayo de 2019 08:21:02 -03 Greg Troxel escribió:
> [snip] 
> > My memory is fuzzy, but I'm pretty sure I''ve seen macs (macos)
> > enter sleep in the middle of a huge compilation.  My impression has
> > been that entering sleep is about timeouts from keyboard/mouse and
> > the implied human.
> > 
> > Systems have all sorts of background processes, like syslog and
> > cron. So this may well be true, but it doesn't seem obviously
> > sensible.
> > 
> > Hence my notion of something specific that can be looked up and the
> > details read about.  
> 
> I also agree it does not strictly depends upon gpsd, so I changed the
> wording to "might keep".
> 
> In fact I would like to know a single case in which this might happen.

I hate sleep and suspend, so I'm the wrong guy to test it.

I tried to google it, but instead I found lots of references to how to
prevent sleeping in Linux...

Can someone that uses sleep test this?

I can't see an obvious difference from the command line:

# gps -n /dev/ttyACM0

# ps -el | fgrep gpsd
F S   UID   PID  PPID  C PRI  NI ADDR SZ WCHAN  TTY          TIME CMD
5 S 65534 26585     1  0  70 -10 -  3512 core_s ?        00:01:07 gpsd

# killall gpsd

# gpsd /dev/ttyACM0

# ps -el | fgrep gpsd
F S   UID   PID  PPID  C PRI  NI ADDR SZ WCHAN  TTY          TIME CMD
5 S 65534 20269     1  0  70 -10 -  1462 core_s ?        00:00:00 gpsd

Both cases the gpsd is in interruptable sleep (S) waiting on "core_s".
Then wakes up periodically to check for inbound connections.  With -n
case also wakes up to check the serial in buffer.

top shows a increase with -n on a RasPi 3 from 0% to 1 or 2% CPU load.

I know any long term process using 2% on my android will get me nags
about sleep being prevented.

On Linux there are vague references to "CPU must be idle" to enter long
term sleep/hibernate.  Unclear if 1% is close enough to idle...

Can someone that uses sleep test this?

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can’t measure it, you can’t improve it." - Lord Kelvin

Attachment: pgptC6b20zqIa.pgp
Description: OpenPGP digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]