Re: minimizing footprint

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

tony37
On 03/24/2013 09:38 AM, [hidden email] wrote:
> It's true, we need learn new tricks, but does not have sense have a service
> not needed on every xo, if we can do it in a better way.

Does this logic apply generally?

12.1.0 has a control panel entry 'Modem configuration'. How many XOs
require this capability?

Tony
_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

Gonzalo Odiard-3


On Sun, Mar 24, 2013 at 11:29 AM, Tony Anderson <[hidden email]> wrote:
On 03/24/2013 09:38 AM, [hidden email] wrote:
It's true, we need learn new tricks, but does not have sense have a service
not needed on every xo, if we can do it in a better way.

Does this logic apply generally?

12.1.0 has a control panel entry 'Modem configuration'. How many XOs require this capability?


Generally. 
In the case we are discussing, we don't need remove a feature, 
only implement it in a different way.

Gonzalo

_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

Paul Fox-2
i think the point john (gilmore) was making is that it's a bad
precedent that simply because a) systemd offers a second-rate API for
scheduling events, and b) we only have a couple of uses for an event
scheduler, that we should therefore switch from a well-established API
to a lousy API.  instead, if systemd can be made to support the cron
API properly, it would be worth making it do so.

barring that, or if it's too much work (likely), then if it saves us
boatloads of disk to make the switch, and our use-cases don't lose
functionality or correctness as a result, then i guess we should
switch.

but when systemd falls out of favor in a year or two we'll just have
to change things again -- either to something new, or back to cron
(which will most certainly still be available).

paul

gonzalo wrote:
 > On Sun, Mar 24, 2013 at 11:29 AM, Tony Anderson <[hidden email]>wrote:
 >
 > > On 03/24/2013 09:38 AM, [hidden email] wrote:
 > >
 > >> It's true, we need learn new tricks, but does not have sense have a
 > >> service
 > >> not needed on every xo, if we can do it in a better way.
 > >>
 > >
 > > Does this logic apply generally?
 > >
 > > 12.1.0 has a control panel entry 'Modem configuration'. How many XOs
 > > require this capability?
 > >
 > >
 > Generally.
 > In the case we are discussing, we don't need remove a feature,
 > only implement it in a different way.
 >
 > Gonzalo
 > part 2     text/plain                 129
 > _______________________________________________
 > Devel mailing list
 > [hidden email]
 > http://lists.laptop.org/listinfo/devel

=---------------------
 paul fox, [hidden email]
_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

Martin Langhoff
Hi Paul,

On Sun, Mar 24, 2013 at 6:34 PM, Paul Fox <[hidden email]> wrote:
> but when systemd falls out of favor in a year or two we'll just have
> to change things again -- either to something new, or back to cron

Not sure whether you're stating that straight or facetiously.

Systemd has some shortcomings, but it seems to be an outstanding step
ahead in Linux system infra. And it is evolving quickly for the better
-- I can't see any fundamental problem with it, and its limitations
and blemishes will be overcome.

cheers,



martin (who's idling on a sunday)
--
 [hidden email]
 -  ask interesting questions
 - don't get distracted with shiny stuff  - working code first
 ~ http://docs.moodle.org/24/en/User:Martin_Langhoff
_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

Gonzalo Odiard-3
In reply to this post by Paul Fox-2
IMHO, if we can speed up the startup time a litle, 
the change worth it. 
We are talking of end users systems, how many kids will add a cron task?
And who propose  the change is dsd, we know him and is
a conservative guy, does not do change without reason.
We did this in the same way the last 30 years, is not the best argument neither...

Gonzalo


On Sun, Mar 24, 2013 at 7:34 PM, Paul Fox <[hidden email]> wrote:
i think the point john (gilmore) was making is that it's a bad
precedent that simply because a) systemd offers a second-rate API for
scheduling events, and b) we only have a couple of uses for an event
scheduler, that we should therefore switch from a well-established API
to a lousy API.  instead, if systemd can be made to support the cron
API properly, it would be worth making it do so.

barring that, or if it's too much work (likely), then if it saves us
boatloads of disk to make the switch, and our use-cases don't lose
functionality or correctness as a result, then i guess we should
switch.

but when systemd falls out of favor in a year or two we'll just have
to change things again -- either to something new, or back to cron
(which will most certainly still be available).

paul

gonzalo wrote:
 > On Sun, Mar 24, 2013 at 11:29 AM, Tony Anderson <[hidden email]>wrote:
 >
 > > On 03/24/2013 09:38 AM, [hidden email] wrote:
 > >
 > >> It's true, we need learn new tricks, but does not have sense have a
 > >> service
 > >> not needed on every xo, if we can do it in a better way.
 > >>
 > >
 > > Does this logic apply generally?
 > >
 > > 12.1.0 has a control panel entry 'Modem configuration'. How many XOs
 > > require this capability?
 > >
 > >
 > Generally.
 > In the case we are discussing, we don't need remove a feature,
 > only implement it in a different way.
 >
 > Gonzalo
 > part 2     text/plain                 129
 > _______________________________________________
 > Devel mailing list
 > [hidden email]
 > http://lists.laptop.org/listinfo/devel

=---------------------
 paul fox, [hidden email]


_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

Paul Fox-2
In reply to this post by Martin Langhoff
martin wrote:
 > Hi Paul,
 >
 > On Sun, Mar 24, 2013 at 6:34 PM, Paul Fox <[hidden email]> wrote:
 > > but when systemd falls out of favor in a year or two we'll just have
 > > to change things again -- either to something new, or back to cron
 >
 > Not sure whether you're stating that straight or facetiously.
 >
 > Systemd has some shortcomings, but it seems to be an outstanding step
 > ahead in Linux system infra. And it is evolving quickly for the better
 > -- I can't see any fundamental problem with it, and its limitations
 > and blemishes will be overcome.

we'll see.

=---------------------
 paul fox, [hidden email]
_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

Daniel Drake-5
In reply to this post by Paul Fox-2
On Sun, Mar 24, 2013 at 4:34 PM, Paul Fox <[hidden email]> wrote:
> i think the point john (gilmore) was making is that it's a bad
> precedent that simply because a) systemd offers a second-rate API for
> scheduling events, and b) we only have a couple of uses for an event
> scheduler, that we should therefore switch from a well-established API
> to a lousy API.  instead, if systemd can be made to support the cron
> API properly, it would be worth making it do so.

systemd's cron-like API is actually more powerful than cron, in that
it can schedule based on seconds precision, and offers monotonic
clocks in addition to calendar times.

It is true that it has not passed the test of time.

> barring that, or if it's too much work (likely), then if it saves us
> boatloads of disk to make the switch, and our use-cases don't lose
> functionality or correctness as a result, then i guess we should
> switch.

It's not a huge saving, but a small amount of work to eliminate a
daemon that runs all the time on every laptop seems worth it to me.

Daniel
_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel
Reply | Threaded
Open this post in threaded view
|

Re: minimizing footprint

John Gilmore
In reply to this post by Gonzalo Odiard-3
> We are talking of end users systems, how many kids will add a cron task?

I was more concerned with other Fedora packages that insert cron jobs
when they are installed.  But if Fedora gets the dependencies right,
rpm would install and run cron at the same time the package that needs
cron is installed, so that's not an issue (unless the package
dependencies are set to assume that cron is always part of the base
system).

The packages you're proposing to change are OLPC-specific packages
anyway, so the changes won't mess up any other Fedora installs.

I do see the point of removing unneeded daemons in every
memory-constrained laptop, and this does look like low hanging fruit.
On the Ubuntu system I happen to be typing on, 'cat /proc/nnnn/status'
shows cron taking 300k of resident RAM and 2.7Mb of virtual memory at
its peak (and I have a place to page out to, which most XO's don't).
The XO system will certainly be more responsive with another 2MB of
memory available for general paging; its worst performance bottleneck
was lack of DRAM when I last looked, and that situation has probably
gotten worse as the system grew.

My thought was that systemd probably wouldn't get much bigger if it
got a parser for cron files.  But your proposed change is less work,
though it only fixes the issue for XO's, not for other
memory-constrained systems.

        John
_______________________________________________
Devel mailing list
[hidden email]
http://lists.laptop.org/listinfo/devel