plplot-devel

[Plplot-devel] The wxwidgets device driver speed/idle time

[Alan W. I. <ir...@be...>]

On 2016-11-14 08:28+0100 Laurent Berger wrote:

> Hi,
>
> I have just made a git pull git.code.sf.net/p/plplot/plplot. Using cmake 
> 3.7.0.rc2 and wxwidgets 3.1.0 and VS 2015. After pressing OK every where, all 
> is fine. thanks for this new version!

Hi Laurent:

I am glad to hear our wxwidgets device driver works for you so well on
Windows (which is the primary development platform used by Phil
Rosenberg for this new wxwidgets device driver).  However, I am having
real trouble with how slow this device is on my Linux platform (Debian
Jessie), and gave a detailed example (last January on this list that
included screen shots of my cpu meter) of the issue which was the cpu
is idle most of the time when communicating data both ways between the
wxwidgets device driver and wxPLViewer application.  However, I don't
know whether this excessivee idle time (compared to all our other
interactive devices available on Linux) is specific to Debian Jessie
or a general issue that needs to be addressed for the Linux
interprocess communication protocol that is being used to communicate
between wxwidgets device driver and wxPLViewer application. So if you
(or anyone else here) has access to a Linux box, some confirmation (or
not) of my (extremely) slow speed results for wxwidgets device driver
and wxPLViewer application would be much appreciated.

Alan

__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-devel] The wxwidgets device driver speed/idle time

[Alan W. I. <ir...@be...>]

Hi Laurent:

I have put my reply to you on list for obvious reasons.

On 2016-11-14 14:26+0100 Laurent Berger wrote:

> Hi Alan,
>
> I would try to help you

Thanks in advance for that.

> but I'm not familiar on linux (see attached file for 
> version).

>From what you included, your Linux platform has access to wxwidgets
version 3.0 (as do I).  But the rest of your platform (old Linux Mint
distribution using a relatively old kernel version 3.2.0-2-amd64
compared to my much more modern Debian Jessie using a fairly
up-to-date kernel version 3.16.0-4-amd64) provides an extremely useful
testing contrast in case the issue is some lameness in the IPC
implementation for my particular Linux kernel.  (IPC, i.e.,
interprocess communication is used to communicate between -dev
wxwidgets and the wxPLViewer application that displays the results.)

> I think it is this message 
> (https://sourceforge.net/p/plplot/mailman/message/34744869/)

Yes, that message and the following one that provided the screenshots referred to
in that message are the relevant ones.

>
> I have build plplot on linux using cmake 3.0.2 using "Unix Makefiles"
> Then sudo make install. I have attached git log result and cmakecache
> Then  I run in folder examples/c++
> ./x01c -dev wxdrivers
>
> I can see a window with 4 curves. Windows title is wxPLviewer. No problem
> Now what do you want to test ?

Speed.

You measure speed on linux using, e.g.,

software@raven> time examples/c/x01c -dev wxwidgets -np
PLplot library version: 5.11.1

real    0m0.652s
user    0m0.064s
sys     0m0.016s
software@raven> 
software@raven> time examples/c/x01c -dev wxwidgets -np
PLplot library version: 5.11.1

real    0m9.327s
user    0m0.072s
sys     0m0.008s

That -np option stands for "no pause" and allows you to measure time
without waiting for the human reaction of clicking on the GUI to
finish the display.

Those two measurements obviously have a huge difference in the real
time used. The first timing is not too bad, but (by chance) the second
timed run shows the problem which is quite often there is a random
drop of at least an order of magnitude and sometimes two orders of
magnitude in speed of communications between -dev wxwidgets and
wxPLviewer due to excessive times when my cpus sit absolutely idle (as
proved above by comparison of the sum of the user + sys times and the
real time required to finish the example in the second time run).
(This excessive idle time is also proved by my screenshots of my cpu
meter that are referred to above.) The above results are for the
current master branch tip.  By way of comparison here are the
consistent timing results I get for -dev xwin, -dev qtwidget (for
Qt5), and -dev xcairo on my platform.

software@raven> time examples/c/x01c -dev xwin -np
PLplot library version: 5.11.1

real    0m0.482s
user    0m0.008s
sys     0m0.008s
software@raven> time examples/c/x01c -dev xwin -np
PLplot library version: 5.11.1

real    0m0.472s
user    0m0.012s
sys     0m0.004s
software@raven> time examples/c/x01c -dev qtwidget -np
PLplot library version: 5.11.1
libGL error: failed to authenticate magic 759
libGL error: failed to load driver: nouveau

real    0m0.472s
user    0m0.088s
sys     0m0.012s
software@raven> 
software@raven> time examples/c/x01c -dev qtwidget -np
PLplot library version: 5.11.1

real    0m0.452s
user    0m0.076s
sys     0m0.020s
software@raven> time examples/c/x01c -dev xcairo -np
PLplot library version: 5.11.1

real    0m0.239s
user    0m0.048s
sys     0m0.012s
software@raven> time examples/c/x01c -dev xcairo -np
PLplot library version: 5.11.1

real    0m0.229s
user    0m0.044s
sys     0m0.012s

If you get consistent timing results for -dev wxwidgets no matter how
often you repeat the experiment and those timing results are within a
factor of two of the timing results for one or more of the above
interactive devices, then it is clear the IPC problem is likely due to
some kernel issue on my Debian Jessie platform, and we can write off
my timing results above to that cause.  But if you also get wildly
inconsistent and often extremely slow wxwidgets results for a
completely different Linux kernel, then either IPC kernel
implementations on Linux are uniformly bad regardless of kernel
version (a conclusion which I doubt) or the way that -dev wxwidgets
and wxPLviewer currently use IPC to communicate between them is set up
in a bad way for the Linux case, that randomly causes excessive idle
times, and that is an issue that the PLplot development team needs to
urgently address if we ever want the wxwidgets device to be popular on
Linux.

I am looking forward to your reply with similar time experiments
on your own Linux platform.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-devel] The wxwidgets device driver speed/idle time

[Laurent B. <lau...@un...>]

Hi Alan,

I haven't got qt installed.
My results looks like yours results :

lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev wxwidgets -np
PLplot library version: 5.11.1
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
103: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
138: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
103: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
138: Having multiple values in <test> isn't supported and may not work 
as expected

real    0m0.407s
user    0m0.108s
sys    0m0.016s
lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev wxwidgets -np
PLplot library version: 5.11.1
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
103: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
138: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
103: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
138: Having multiple values in <test> isn't supported and may not work 
as expected

real    0m10.997s
user    0m0.064s
sys    0m0.040s
lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xwin -np
PLplot library version: 5.11.1

real    0m0.122s
user    0m0.056s
sys    0m0.020s
lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xwin -np
PLplot library version: 5.11.1

real    0m0.122s
user    0m0.060s
sys    0m0.016s
lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xcairo -np
PLplot library version: 5.11.1
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
103: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
138: Having multiple values in <test> isn't supported and may not work 
as expected

real    0m0.125s
user    0m0.088s
sys    0m0.016s
lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xcairo -np
PLplot library version: 5.11.1
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
103: Having multiple values in <test> isn't supported and may not work 
as expected
Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 
138: Having multiple values in <test> isn't supported and may not work 
as expected

real    0m0.128s
user    0m0.084s
sys    0m0.020s
lb@mp12-lb:~/plplot.git/examples/c++$






Le 14/11/2016 à 22:19, Alan W. Irwin a écrit :
> Hi Laurent:
>
> I have put my reply to you on list for obvious reasons.
>
> On 2016-11-14 14:26+0100 Laurent Berger wrote:
>
>> Hi Alan,
>>
>> I would try to help you
>
> Thanks in advance for that.
>
>> but I'm not familiar on linux (see attached file for version).
>
> From what you included, your Linux platform has access to wxwidgets
> version 3.0 (as do I).  But the rest of your platform (old Linux Mint
> distribution using a relatively old kernel version 3.2.0-2-amd64
> compared to my much more modern Debian Jessie using a fairly
> up-to-date kernel version 3.16.0-4-amd64) provides an extremely useful
> testing contrast in case the issue is some lameness in the IPC
> implementation for my particular Linux kernel.  (IPC, i.e.,
> interprocess communication is used to communicate between -dev
> wxwidgets and the wxPLViewer application that displays the results.)
>
>> I think it is this message 
>> (https://sourceforge.net/p/plplot/mailman/message/34744869/)
>
> Yes, that message and the following one that provided the screenshots 
> referred to
> in that message are the relevant ones.
>
>>
>> I have build plplot on linux using cmake 3.0.2 using "Unix Makefiles"
>> Then sudo make install. I have attached git log result and cmakecache
>> Then  I run in folder examples/c++
>> ./x01c -dev wxdrivers
>>
>> I can see a window with 4 curves. Windows title is wxPLviewer. No 
>> problem
>> Now what do you want to test ?
>
> Speed.
>
> You measure speed on linux using, e.g.,
>
> software@raven> time examples/c/x01c -dev wxwidgets -np
> PLplot library version: 5.11.1
>
> real    0m0.652s
> user    0m0.064s
> sys     0m0.016s
> software@raven> software@raven> time examples/c/x01c -dev wxwidgets -np
> PLplot library version: 5.11.1
>
> real    0m9.327s
> user    0m0.072s
> sys     0m0.008s
>
> That -np option stands for "no pause" and allows you to measure time
> without waiting for the human reaction of clicking on the GUI to
> finish the display.
>
> Those two measurements obviously have a huge difference in the real
> time used. The first timing is not too bad, but (by chance) the second
> timed run shows the problem which is quite often there is a random
> drop of at least an order of magnitude and sometimes two orders of
> magnitude in speed of communications between -dev wxwidgets and
> wxPLviewer due to excessive times when my cpus sit absolutely idle (as
> proved above by comparison of the sum of the user + sys times and the
> real time required to finish the example in the second time run).
> (This excessive idle time is also proved by my screenshots of my cpu
> meter that are referred to above.) The above results are for the
> current master branch tip.  By way of comparison here are the
> consistent timing results I get for -dev xwin, -dev qtwidget (for
> Qt5), and -dev xcairo on my platform.
>
> software@raven> time examples/c/x01c -dev xwin -np
> PLplot library version: 5.11.1
>
> real    0m0.482s
> user    0m0.008s
> sys     0m0.008s
> software@raven> time examples/c/x01c -dev xwin -np
> PLplot library version: 5.11.1
>
> real    0m0.472s
> user    0m0.012s
> sys     0m0.004s
> software@raven> time examples/c/x01c -dev qtwidget -np
> PLplot library version: 5.11.1
> libGL error: failed to authenticate magic 759
> libGL error: failed to load driver: nouveau
>
> real    0m0.472s
> user    0m0.088s
> sys     0m0.012s
> software@raven> software@raven> time examples/c/x01c -dev qtwidget -np
> PLplot library version: 5.11.1
>
> real    0m0.452s
> user    0m0.076s
> sys     0m0.020s
> software@raven> time examples/c/x01c -dev xcairo -np
> PLplot library version: 5.11.1
>
> real    0m0.239s
> user    0m0.048s
> sys     0m0.012s
> software@raven> time examples/c/x01c -dev xcairo -np
> PLplot library version: 5.11.1
>
> real    0m0.229s
> user    0m0.044s
> sys     0m0.012s
>
> If you get consistent timing results for -dev wxwidgets no matter how
> often you repeat the experiment and those timing results are within a
> factor of two of the timing results for one or more of the above
> interactive devices, then it is clear the IPC problem is likely due to
> some kernel issue on my Debian Jessie platform, and we can write off
> my timing results above to that cause.  But if you also get wildly
> inconsistent and often extremely slow wxwidgets results for a
> completely different Linux kernel, then either IPC kernel
> implementations on Linux are uniformly bad regardless of kernel
> version (a conclusion which I doubt) or the way that -dev wxwidgets
> and wxPLviewer currently use IPC to communicate between them is set up
> in a bad way for the Linux case, that randomly causes excessive idle
> times, and that is an issue that the PLplot development team needs to
> urgently address if we ever want the wxwidgets device to be popular on
> Linux.
>
> I am looking forward to your reply with similar time experiments
> on your own Linux platform.
>
> Alan
> __________________________
> Alan W. Irwin
>
> Astronomical research affiliation with Department of Physics and 
> Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
>
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
>
> Linux-powered Science
> __________________________


-- 
Laurent Berger
Enseignant-chercheur I.U.T. du Mans
    Département Gestion des Entreprises et des Administrations
    Département Mesures Physiques
Tél 02 43 83 37 09

Re: [Plplot-devel] The wxwidgets device driver speed/idle time

[Alan W. I. <ir...@be...>]

On 2016-11-15 14:59+0100 Laurent Berger wrote:

> Hi Alan,
>
> I haven't got qt installed.
> My results looks like yours results :
>
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev wxwidgets -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m0.407s
> user    0m0.108s
> sys    0m0.016s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev wxwidgets -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m10.997s
> user    0m0.064s
> sys    0m0.040s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xwin -np
> PLplot library version: 5.11.1
>
> real    0m0.122s
> user    0m0.056s
> sys    0m0.020s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xwin -np
> PLplot library version: 5.11.1
>
> real    0m0.122s
> user    0m0.060s
> sys    0m0.016s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xcairo -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m0.125s
> user    0m0.088s
> sys    0m0.016s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xcairo -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m0.128s
> user    0m0.084s
> sys    0m0.020s
> lb@mp12-lb:~/plplot.git/examples/c++$

Hi Laurent:

Thanks very much for the above report.

Your timing results verify mine and show that -dev wxwidgets sometimes
has very poor "real" timing results even though the actual cpu times
indicated by the sum of "user" + "sys" are quite fast and consistent.

For this case

> real    0m10.997s
> user    0m0.064s
> sys    0m0.040s

the sum of user + sys is 0.104 seconds so that means for this test the
cpu was idle for 10.893 seconds out of the 10.997 seconds or 99.05 per
cent of the time which is equivalent to two (!) orders of magnitude
slowdown.  Since your results and mine confirm the problem for a wide
range of Linux kernel versions, my working hypothesis is these
excessive idle times/slowdowns are due to the inefficient way that
Linux kernel IPC is being used to implement the necessary two-way
communication between -dev wxwidgets and the wxPLviewer application on
Linux.

Gaining two orders of magnitude in speed is certainly an attractive
goal for the wxwidgets device/wxPLviewer combination so I am tempted
to immediately start reviewing the Linux kernel IPC use between -dev
wxwidgets and wxPLviewer.  However, my current knowledge of Linux
kernel IPC is small, and I have many other PLplot issues on my plate
at the moment so I will have to delay this for quite a while.  Thus,
there is an opportunity for someone else to contribute substantially
here if they have some knowledge about the most efficient way to use
Linux kernel IPC or want to learn more about that interesting topic.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-devel] The wxwidgets device driver speed/idle time

[<p.d...@gm...>]

Hi Alan, Laurent
I can't say I have much to add. It wasn't obvious to me where the lag was last time i looked, but I agree it is an issue

I haven't done much plplot work for some time, but when I can i will look at this. If anyone has any suggestions I would be keen to hear.

Phil 

Sent from my Windows 10 phone

From: Alan W. Irwin
Sent: 15 November 2016 21:22
To: Laurent Berger
Cc: PLplot development list
Subject: Re: [Plplot-devel] The wxwidgets device driver speed/idle time

On 2016-11-15 14:59+0100 Laurent Berger wrote:

> Hi Alan,
>
> I haven't got qt installed.
> My results looks like yours results :
>
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev wxwidgets -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m0.407s
> user    0m0.108s
> sys    0m0.016s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev wxwidgets -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m10.997s
> user    0m0.064s
> sys    0m0.040s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xwin -np
> PLplot library version: 5.11.1
>
> real    0m0.122s
> user    0m0.056s
> sys    0m0.020s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xwin -np
> PLplot library version: 5.11.1
>
> real    0m0.122s
> user    0m0.060s
> sys    0m0.016s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xcairo -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m0.125s
> user    0m0.088s
> sys    0m0.016s
> lb@mp12-lb:~/plplot.git/examples/c++$ time ./x01 -dev xcairo -np
> PLplot library version: 5.11.1
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 103: 
> Having multiple values in <test> isn't supported and may not work as expected
> Fontconfig warning: "/etc/fonts/conf.d/65-droid-sans-fonts.conf", line 138: 
> Having multiple values in <test> isn't supported and may not work as expected
>
> real    0m0.128s
> user    0m0.084s
> sys    0m0.020s
> lb@mp12-lb:~/plplot.git/examples/c++$

Hi Laurent:

Thanks very much for the above report.

Your timing results verify mine and show that -dev wxwidgets sometimes
has very poor "real" timing results even though the actual cpu times
indicated by the sum of "user" + "sys" are quite fast and consistent.

For this case

> real    0m10.997s
> user    0m0.064s
> sys    0m0.040s

the sum of user + sys is 0.104 seconds so that means for this test the
cpu was idle for 10.893 seconds out of the 10.997 seconds or 99.05 per
cent of the time which is equivalent to two (!) orders of magnitude
slowdown.  Since your results and mine confirm the problem for a wide
range of Linux kernel versions, my working hypothesis is these
excessive idle times/slowdowns are due to the inefficient way that
Linux kernel IPC is being used to implement the necessary two-way
communication between -dev wxwidgets and the wxPLviewer application on
Linux.

Gaining two orders of magnitude in speed is certainly an attractive
goal for the wxwidgets device/wxPLviewer combination so I am tempted
to immediately start reviewing the Linux kernel IPC use between -dev
wxwidgets and wxPLviewer.  However, my current knowledge of Linux
kernel IPC is small, and I have many other PLplot issues on my plate
at the moment so I will have to delay this for quite a while.  Thus,
there is an opportunity for someone else to contribute substantially
here if they have some knowledge about the most efficient way to use
Linux kernel IPC or want to learn more about that interesting topic.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

------------------------------------------------------------------------------
_______________________________________________
Plplot-devel mailing list
Plp...@li...
https://lists.sourceforge.net/lists/listinfo/plplot-devel

Re: [Plplot-devel] The wxwidgets device driver speed/idle time

[Alan W. I. <ir...@be...>]

On 2016-11-15 22:52-0000 p.d...@gm... wrote:

> Hi Alan, Laurent
> I can't say I have much to add. It wasn't obvious to me where the lag was last time i looked, but I agree it is an issue
>

> I haven't done much plplot work for some time, but when I can i will
look at this. If anyone has any suggestions I would be keen to hear.

Hi Phil:

>From what you say above it sounds like any serious time you can spend
on this would be pretty limited and not immediate.  Given those
conditions, I think that likely the most valuable thing you can do for
us immediately is to write up a paragraph or so describing how you are
currently using IPC to communicate between -dev wxwidgets and
wxPLViewer including a summary of where waits might be occurring. I
suggest you commit that overview in the new file
drivers/README.wxwidgets_IPC (or I will do that if you just want to
send your overview by e-mail). That IPC overview would be a big help
in getting up to speed for anyone else wanting to figure out what is
going on with these weird idle times that we are encountering on the
Linux platform.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-devel] The wxwidgets device driver speed/idle time

[Alan W. I. <ir...@be...>]

Hi Phil:

I got tired of speculating about more efficient Linux IPC methods so I
implemented (as of commit e166866) the "Unnamed semaphores example"
paradigm outlined starting on page 73 of
<http://man7.org/conf/lca2013/IPC_Overview-LCA-2013-printable.pdf>. (I
discussed this possibility in earlier versions of this thread.)
This test project is described in cmake/test_linux_ipc/README.  Follow
the directions there to test it on your own Linux systems.

My tests show this method is quite efficient for transferring data
between processes on POSIX-compliant systems.  The transfer process is
under complete control of unnamed semaphores that reside in the shared
memory so coordinating required data transfer between two applications
is completely straightforward, and there is very little idle time
required during these transfers.

The IPC-relevant routines that are called in this test project are
shm_open, ftruncate, mmap, sem_init, sem_wait, sem_post, and
shm_unlink, and according to the Linux man pages, all of these
functions are provided by platforms that are compliant with POSIX.1-2001.

So on the present evidence from this test project I think this method
looks extremely promising for efficiently transferring data both ways
between -dev wxwidgets and wxPLViewer for POSIX-compliant systems.

If after trying the test of the method documented in 
cmake/test_linux_ipc/README you agree with this assessment, would you
be willing to have a go at the required wxwidgets and wxPLViewer
source code changes to see if that essentially eliminates the
large idle times we have now on Linux?

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-devel] The wxwidgets device driver speed/idle time

[Phil R. <p.d...@gm...>]

Hi Alan
It's on my list - I will do it as soon as I can :-). I actually think
I had a similar system like this set up - although I think I used
named semaphores. I would have to go back to the code to check.

Phil

On 19 November 2016 at 07:39, Alan W. Irwin <ir...@be...> wrote:
> Hi Phil:
>
> I got tired of speculating about more efficient Linux IPC methods so I
> implemented (as of commit e166866) the "Unnamed semaphores example"
> paradigm outlined starting on page 73 of
> <http://man7.org/conf/lca2013/IPC_Overview-LCA-2013-printable.pdf>. (I
> discussed this possibility in earlier versions of this thread.)
> This test project is described in cmake/test_linux_ipc/README.  Follow
> the directions there to test it on your own Linux systems.
>
> My tests show this method is quite efficient for transferring data
> between processes on POSIX-compliant systems.  The transfer process is
> under complete control of unnamed semaphores that reside in the shared
> memory so coordinating required data transfer between two applications
> is completely straightforward, and there is very little idle time
> required during these transfers.
>
> The IPC-relevant routines that are called in this test project are
> shm_open, ftruncate, mmap, sem_init, sem_wait, sem_post, and
> shm_unlink, and according to the Linux man pages, all of these
> functions are provided by platforms that are compliant with POSIX.1-2001.
>
> So on the present evidence from this test project I think this method
> looks extremely promising for efficiently transferring data both ways
> between -dev wxwidgets and wxPLViewer for POSIX-compliant systems.
>
> If after trying the test of the method documented in
> cmake/test_linux_ipc/README you agree with this assessment, would you
> be willing to have a go at the required wxwidgets and wxPLViewer
> source code changes to see if that essentially eliminates the
> large idle times we have now on Linux?
>
>
> Alan
> __________________________
> Alan W. Irwin
>
> Astronomical research affiliation with Department of Physics and Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
>
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
>
> Linux-powered Science
> __________________________