I just posted to the Xenomai mailing list reporting some puzzling behaviour
with respect to latency spikes.
There's some good feedback and advice coming in:
http://www.xenomai.org/pipermail/xenomai/2013-February/027675.html
in particular, give the "nohlt idle=poll" boot option a try; see Henri
On 13 February 2013 10:15, Michael Haberler wrote:
> http://www.xenomai.org/pipermail/xenomai/2013-February/027675.html
This sounds familiar, there is a similar situation with RTAI:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?RealTime
and the "cpu hog" near the bottom.
After a bit more searching,
Am 13.02.2013 um 11:37 schrieb andy pugh:
> On 13 February 2013 10:15, Michael Haberler wrote:
>
>> http://www.xenomai.org/pipermail/xenomai/2013-February/027675.html
>
> This sounds familiar, there is a similar situation with RTAI:
> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?RealTime
> and the
Same here...
http://imagebin.org/246546
That is after adding nohlt idle=poll to the boot options. This is the
machine we have been testing on - 100+us latency unless there is some
sort of do-nothing process running.
it runs a bit hotter..
sam
On 2/13/2013 5:08 AM, Michael Haberler wrote:
>
On Wed, Feb 13, 2013 at 8:43 AM, sam sokolik wrote:
>
> it runs a bit hotter..
>
> That's what they were saying on the xenomai list. I guess if you aren't
running it all the time on an inadequate power supply it should be ok. I
assume this is what the busy loop that has been mentioned many time
On 2/13/2013 6:08 AM, Michael Haberler wrote:
> Am 13.02.2013 um 11:37 schrieb andy pugh:
>
>> On 13 February 2013 10:15, Michael Haberler wrote:
>>
>>> http://www.xenomai.org/pipermail/xenomai/2013-February/027675.html
>> This sounds familiar, there is a similar situation with RTAI:
>> http://wik
for reference - from
http://www.kernel.org/doc/Documentation/kernel-parameters.txt
nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
wfi(ARM) instruction doesn't work correctly and not to
use it. This is also useful when
On Wed, Feb 13, 2013 at 12:29 PM, Michael Haberler wrote:
> the hlt instruction doesn't work correctly
>
It bugs me that in these days of high performance and energy aware
computers that the hlt command apparently doesn't work properly on a lot of
processors. This affects realtime users the most
Am 13.02.2013 um 18:47 schrieb Eric Keller:
> On Wed, Feb 13, 2013 at 12:29 PM, Michael Haberler wrote:
>
>> the hlt instruction doesn't work correctly
>>
> It bugs me that in these days of high performance and energy aware
> computers that the hlt command apparently doesn't work properly on a
Am 13.02.2013 um 18:52 schrieb Michael Haberler:
>
> well I dont know enough about how the electronics of all this is implemented,
> but if any of the power supply (like onboard SMPS regulators) is involved in
> the CPU voltage reduction of some of these C* states, I wouldnt be suprised
> that
Michael
On a quick test on the Q8200 Intel quad core
no-hlt idle=mwait - hangs similar to kernel panic
no-hlt idle=nomwait - 18K base jitter - approx same as with no params
no-hlt idle=poll - 10K base jitter
nohlt idle-poll - 11K
Loading was the same, 4 x glxgears and Firefox open and download
Last one should of course read
nohlt idle=poll - 11K
On 13/02/13 18:11, Schooner wrote:
> Michael
>
> On a quick test on the Q8200 Intel quad core
>
> no-hlt idle=mwait - hangs similar to kernel panic
> no-hlt idle=nomwait - 18K base jitter - approx same as with no params
> no-hlt idle=poll - 10
Bugs item #3604558, was opened at 2013-02-13 10:45
Message generated for change (Tracker Item Submitted) made by jeromepilot22
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=106744&aid=3604558&group_id=6744
Please note that this message will contain a full copy of
Bugs item #3604558, was opened at 2013-02-13 10:45
Message generated for change (Comment added) made by seb_kuzminsky
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=106744&aid=3604558&group_id=6744
Please note that this message will contain a full copy of the comme
For this system (amd) the only thing that seems to make a difference is
the idle=poll (no-halt seems to have no effect)
I go from 200+us latency to 7us top.
sam
On 2/13/2013 11:29 AM, Michael Haberler wrote:
> for reference - from
> http://www.kernel.org/doc/Documentation/kernel-parameters.txt
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 2/13/2013 11:52 AM, Michael Haberler wrote:
>
> Am 13.02.2013 um 18:47 schrieb Eric Keller:
>
>> On Wed, Feb 13, 2013 at 12:29 PM, Michael Haberler
>> wrote:
>>
>>> the hlt instruction doesn't work correctly
The hlt instruction works fine, it ju
I spent the day playing with this on the Intel Atom D2700MUD. I rebuilt the
system using the rtos-integration-preview3-merged-into-master
branch. I rebuilt it for another reason (stupid graphics fubar) and figured I
would try this branch for the heck of it.
Neither no-halt nor idle=poll had an
On 2/13/2013 10:28 PM, Tom Easterday wrote:
> I spent the day playing with this on the Intel Atom D2700MUD. I rebuilt the
> system using the rtos-integration-preview3-merged-into-master
> branch. I rebuilt it for another reason (stupid graphics fubar) and figured
> I would try this branch for t
Hello emc-dev and xenomai lists,
I'm very pleased to announce the availability of one-size-fits-all
Xenomai kernel packages for x86 arch. These packages have been
successfully tested on umpteen motherboards at this point, and are ready
for wider testing.
* Beta status *
Distros: Ubuntu Precise a
While working on an offset display widget, I notice that it is quite possible
for
linuxcnc to _not_ update the var file after say a MDI command G10 L2 P0 etc.
I notice reloading the Gcode file forces this update.
Is there any way to force linuxcnc directly to update the var file, preferably
in
20 matches
Mail list logo
