Re: [Emc-developers] RTAI Testing

[andy pugh]

On Fri, 15 May 2020 at 00:27, andy pugh  wrote:

> > I am going to halt that test and run the Seb abs.0 test with
> > RTAI-uspace to see if that is different to RTAI-kernel. (ie
> > LXRT-realtime not RTAI-realtime)
>
> Which got to 10,000 cycles, but then the kernel-mode RTAI test often does.
> I will leave it overnight.

Failed at 12,877 cycles after slightly more than 12 hours running.
(the LXRT-realtime test seems to take rather longer to start)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers

Re: [Emc-developers] RTAI Testing

[andy pugh]

On Thu, 14 May 2020 at 12:21, andy pugh  wrote:
>
> Trying out Jepler's test:
> http://tom-itx.no-ip.biz:81/~tom-itx/irc/logs/%23linuxcnc-devel/2020-05-07.html
> On real hardware I am currently at 114,000 cycles with no faults.
> So the issue seen _might_ be QEMU related.
>
> I am going to halt that test and run the Seb abs.0 test with
> RTAI-uspace to see if that is different to RTAI-kernel. (ie
> LXRT-realtime not RTAI-realtime)

Which got to 10,000 cycles, but then the kernel-mode RTAI test often does.
I will leave it overnight.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers

Re: [Emc-developers] RTAI Testing

[Gene Heskett]

On Thursday 14 May 2020 07:21:51 andy pugh wrote:

> Trying out Jepler's test:
> http://tom-itx.no-ip.biz:81/~tom-itx/irc/logs/%23linuxcnc-devel/2020-0
>5-07.html On real hardware I am currently at 114,000 cycles with no
> faults. So the issue seen _might_ be QEMU related.
>
Thats a 4.14 kernel, which on an rpi4 is not fully aware of the new video 
modes available on the pi's.

When I asked raspbian about a realtime preempt, I was shown a 4.19 
version that was supplied as a src file, but it came with zero support, 
questions would not be answered.  And weren't.  So I had to make quite a 
few passes at menuconfig to build it, and its still building stuff 
foreign to the pi, but it built.

Then I figured out a way, rather than making a huge deb out of it, to 
make a 25.5 meg tarball that only overwrote the std kernel with the 
preempt-rt part, ditto the stuff in /lib/modules which booted my pi and 
gives a latency time plenty good enough for uspace, 15 u-secs, up to 
around 50 if the pi is real busy, like browsing the net with firefox. 
Its dead stable and runs the pi4 nicely.  With nice speedy 60 fps video. 

Power isn't 100% here, so I put a $35 Cyberpower ups on it, and have only 
rebooted it once since, the upgrade to buster 10.4 gave it a tummy ache 
but the reboot was clean and still is.

It's available from this machine but is a 2.6GB dl at my dsl speeds and 
would need a reconfig session for wintel stuff as it presently making 
armhf.  It might be worth looking at for uspace builds.

The bare 25,415,212 byte Feb 24 11:09 kernel-rt.tgz is there if you want 
to try it on the buildbot, follow the README.txt to install it.

To find either, or both, add lathe-stf to the browser address bar of the 
link in my sig, then click on linuxcnc4rpi4, and you should be there.


> I am going to halt that test and run the Seb abs.0 test with
> RTAI-uspace to see if that is different to RTAI-kernel. (ie
> LXRT-realtime not RTAI-realtime)


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


___
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers