Theo Veenker wrote:
> Hi all,
>
> I'm preparing a Toshiba Satellite S1400-153E laptop for field work.
> It used to run redhat 9 + RTLinux, but now it runs Ubuntu 6.06 +
> Xenomai 2.3.
>
> Unfortunately the latency test shows high latencies (200..500us) at
> irregular intervals (5..20 sec) whereas
() because I just want to exchange simple integer values.
Thanks for help,
Markus Franke
Gilles Chanteperdrix wrote:
> Gilles Chanteperdrix wrote:
> > Markus Franke wrote:
> > > OK. And if there is an overflow is there anything I can do during
> > > runtime, e.g. ju
OK. And if there is an overflow is there anything I can do during
runtime, e.g. just to overwrite the data in the queue?
Regards,
Markus Franke
Gilles Chanteperdrix wrote:
> Markus Franke wrote:
>
>>what happens if rt_pipe_write() returns
>>-ENOMEM? Does the data to be writt
Gilles Chanteperdrix wrote:
> The fourth argument of rt_pipe_create is the size of the reserved pool
> used by the pipe. If you pass 0, the system heap is used. The size of
> the system heap is configured via the kernel configuration.
>
> rt_pipe_alloc allocates a message from the used pool (the r
happens if I
just ignore this error code? Does it simply mean that values are dropped
until the first value is read from the pipe by the User-Space Task? If
so, how can I increase the size of the message pipe? rt_pipe_alloc()
doesn't seem to solve the problem, isn't it?
Thanks in advance
/Xenomai_IRQ_Handler.png.
Where do the discrete distances between the results come from? It looks
as if it is always around 1 us. Why aren't the results distributed
continuously? By the way, my bucket_size wasn't 1 us, rather it was 1
ns, so actually without any buckets.
Regards,
Mar
and lp were loaded during
system boot. When I removed these modules in order to load my parport
module, the parallelport was always switched off by linux and it was not
possible to activate it again.
>
> Thanks
>
> Roland
>
>> the kernel commandline.
>> Maybe this can a
Wolfgang Grandegger wrote:
> Good hint. Roland, could you try that first. I think this does not
> require the parport module loaded, doesn't it?
No, it doesn't need the parport module nor the lp module beeing loaded.
begin:vcard
fn:Markus Franke
n:Franke;Markus
adr;quoted-printable:;;Vettersstra=C
f"-option to
the kernel commandline.
Maybe this can also help you.
Regards,
Markus Franke
roland Tollenaar wrote:
>> >
>> > FATAL: Error inserting xeno_can_peak_dng (xeno_can_peak_dng.ko): No
>> such
>> > device
>>
>> Is there som additional kernel
Dmitry Adamushko wrote:
> On 14/02/07, Markus Franke <[EMAIL PROTECTED]> wrote:
>
>> Jan Kiszka wrote:
>> > Of course, irqloop runs in *primary* mode to be able to handle the
>> > events deterministically. So it is not directly affected by
>> CONFIG_PREE
Jan Kiszka wrote:
> Of course, irqloop runs in *primary* mode to be able to handle the
> events deterministically. So it is not directly affected by CONFIG_PREEMPT.
If I start irqloop in User-Mode, a thread is simply created via Linux
system call pthread_create() which interacts with the
xeno_irqb
equency of 1 ms. This gives 6 samples. Is this too
less in order to exclude noise?
Regards,
Markus Franke
begin:vcard
fn:Markus Franke
n:Franke;Markus
adr;quoted-printable:;;Vettersstra=C3=9Fe 64/722;Chemnitz;Saxony;09126;Germany
email;internet:[EMAIL PROTECTED]
x-mozilla-html:FALSE
url:http:
Jan Kiszka wrote:
> Markus Franke wrote:
>>I am running some latency tests with irqbench/irqloop. I am wondering
>>whether it would be possible to achieve better results when activating
>>CONFIG_PREEMPT and CONFIG_PREEMPT_VOLUNTARILY during the kernel
>>configuration
;pingflood" and "cpuburn". Nevertheless, I can only achieve worse
results when activating CONFIG_PREEMPT.
Thanks in advance and regards,
Markus Franke
___
Xenomai-help mailing list
Xenomai-help@gna.org
https://mail.gna.org/listinfo/xenomai-help
Jan Kiszka wrote:
> This test mode exploits the I-pipe support for multiple domains
> intensively.
I knowI read the nice txt-document :-)
> By default, Xenomai is configured to be the topmost domain
> and is optimised for this use case. But it can also be "second-best" and
> let someone else
temporarily unavailable
---snap---
This occurs somewhere when calling:
ioctl(benchdev, RTTST_RTIOC_IRQBENCH_START, &config)
Any quick suggestions how to solve this problem?
By the way: I read already the provided document in "txt" folder.
Thanks in advance,
Markus Fran
Preetam Joshi wrote:
> But i get the error as /usr/bin/ld: cannot find -lnative
> collect2: ld returned 1 exit status
>
> How do i resolve this error.
You should maybe add something like "-L/lib" to your
LDFLAGS in the Makefile. The problem is that "libnative.a" is not found
during linking proces
Well, some symbols which are referenced by your module are not found in
the kernel. A "dmesg()"-call should provide you some information about
the missing symbols. Maybe you missed inserting the "RTDM"-kernel module?
Regards,
Markus
mani bhatti wrote:
> Hi all
> i have a problem inserting the rea
Anders Blomdell wrote:
> The code looks OK, but if you hit some unsigned - signed - unsigned
> conversion during wraparounds, the latencies experienced suggest that
> you run a 1.5 GHz computer (or possibly a 3GHz, never get the 2^32 vs
> 2^31 stuff right):
>
> 2^31/1.5 -> 1.4317e+09 Hz
>
> BTW
worst case.
Thanks in advance,
Markus Franke
begin:vcard
fn:Markus Franke
n:Franke;Markus
adr;quoted-printable:;;Vettersstra=C3=9Fe 64/722;Chemnitz;Saxony;09126;Germany
email;internet:[EMAIL PROTECTED]
x-mozilla-html:FALSE
url:http://www.tu-chemnitz.de/~franm
d be used for developing hard
realtime drivers. But I just need to measure latencies and I don't want
to develop a huge realtime driver. I just need some values for
comparison with the driver working on plain linux. I was wondering
whether it is possible to rewrite the driver working on plain linu
just a small problem which can be fixed easily.
The kernel-module as well as the user-space task are attached.
Thanks in advance and regards,
Markus Franke
PS.: I am aware of the fact that there is already a testcase in the
testsuite for measuring external interrupt latencies. Nevertheless, I
would
are delegated to my ISR? Do I
need to create an extra realtime task via "rt_task_create()" in order to
achieve hard realtime response, or is it enough to install the
interrupt-handler from module_init().
Thanks for help,
Markus Franke
begin:vcard
fn:Markus Franke
n:Franke;Markus
adr;quo
Xenomai. I just want to have some small
hints how to start developing.
Thanks in advance,
Markus Franke
begin:vcard
fn:Markus Franke
n:Franke;Markus
adr;quoted-printable:;;Vettersstra=C3=9Fe 64/722;Chemnitz;Saxony;09126;Germany
email;internet:[EMAIL PROTECTED]
x-mozilla-html:FALSE
url:http://www.tu
y of the linux-kernel
increase when using CONFIG_PREEMPT, the time to the next preemption
point should decrease, right? I don't get the point of your last sentence.
Thanks for help,
Markus Franke
--
Nichts ist so praktisch wie eine gute Theorie!
begin:vcard
fn:Markus Franke
n:Franke;Marku
Dear Xenomai Users,
I want to measure interrupt latencies and I guess the two test-programs
mentioned in the subject are suitable for doing so. I wonder whether
there is any documentation about these two tests and especially how to
use them (hardware requirements, etc.)
Thanks in advance,
Markus
are still
at least 5% above the non-CONFIG_PREEMPT values. I did several testruns
and the result were reobservable. How much improvments with additional
fine granular Kernel Preemptability can one expect?
Thanks in advance,
Markus Franke
begin:vcard
fn:Markus Franke
n:Franke;Markus
adr;quot
the CONFIG_PREEMPT - Case.
The complete Test-Code is attached.
My System is:
Linux Kernel 2.6.17
Adeos Patch: adeos-ipipe-2.6.17-i386-1.5-00.patch
Xenomai 2.2.4
Thanks for help,
Markus Franke
---
#include
#include
#include
#include
#include
#include
#include
#include
#include
#in
28 matches
Mail list logo