Hello every one:
I have a problem on x86 smp system .my linux is 3.10.32 and 4.1.18
with xenomai patch xenomai-2.6.4,xenomai-2.6.5 .
when grub set maxcpus=1 the system run fine. but when set max cpus=2(my cpu is
dual-core ),the error is blow .
My system run realtime task and large
What was the worst-case latency with HT enabled?
The largest value I've observed was 883 microseconds, which happend only
once.
Typical values range around 120-150 microseconds. When running the
stress test it goes < 20 microsecond.
Am 1.3.18 um 12:53 PM schrieb Philippe Gerum:
On 03/01/20
On 03/01/2018 06:46 PM, Julian Viereck wrote:
>> 1) disable hyperthreading and running a single core
> Seems like the latency problem is due to enabled hyperthreading.
>
> I reran the latency tests with hyperthreading disabled and the latencies
> stay below 20 microseconds now. This is with 1 and
On 02/21/2018 07:33 PM, Jan Kiszka wrote:
> Hi,
>
> likely due to some platform oddities (still has to be analyzed in more
> details), we are running into the situation that xntimer_grab_hardware
> on a x86-64 host tunes nkclock.wallclock_offset to a value that is not
> compatible with the 32-bit
1) disable hyperthreading and running a single core
Seems like the latency problem is due to enabled hyperthreading.
I reran the latency tests with hyperthreading disabled and the latencies
stay below 20 microseconds now. This is with 1 and 8 cores. Sorry for
reporting something different yest
Are you using Xenomai 2.6 or 3? If you are using 3 use this guide instead:
https://xenomai.org/documentation/xenomai-3/html/xeno3prm/group__alchemy__timer.html
Post back if nothing there can help, but you may want to loot at the
source for some of the utilities that calculate latency as a exampl
Hi,
I have one question about Xenomai native API
1) What is the native API equivalent for clock_gettime() ?
I have seen this Xenomai native API guide but I could not find it.
https://xenomai.org/documentation/xenomai-2.6/pdf/native-api.pdf
I have a simple pthread based POSIX application which I
On 02/28/2018 08:36 PM, Jan Kiszka wrote:
> On 2018-02-28 19:24, Philippe Gerum wrote:
>>
>> Gentlemen, we have 30 patches waiting for review by RTnet gurus in
>> wip/rtnet-fixes. We already had some positive feedback about them fixing
>> the ugly SMAP-related issue, which we may want to merge to s
On Wed, Feb 28, 2018 at 8:19 PM, Greg Gallagher wrote:
> What's your prototype for dummy_write? Can you post the function
> definition. It looks like there's a rt_write is expecting a different
> type.
>
OK it is working now after adding "const" to void *buf in write function.
Thank you for the
On 02/28/2018 10:40 PM, Julian Viereck wrote:
> Thanks for your pointer to hwlatdetect and possible SMI issues.
>
> It looks like the hwlatdetect script is not working with the latest
> kernel anymore. However, it appears the hwlat is now part of the
> kernel's ftrace [1]. Here is the kernel docum
10 matches
Mail list logo
