Eric H. Johnson wrote:
> I think I found something. First I manually applied all of the pin
> assignments (net commands). Nothing caused a run-away. Then I started
> playing with the parameters, specifically position-scale. If I set
> position-scale to a positive value, 2000 in my case, then everyt
Sebastian,
I think I found something. First I manually applied all of the pin
assignments (net commands). Nothing caused a run-away. Then I started
playing with the parameters, specifically position-scale. If I set
position-scale to a positive value, 2000 in my case, then everything appears
to wor
Feature Requests item #2117197, was opened at 2008-09-17 23:11
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=356744&aid=2117197&group_id=6744
Please note that this message will contain a
Steve, Sebastian,
So far the stepgens are working. But all I have done is load hostmot2 and
hm2_5i20. I linked a signal to the enable for stepper 1, and set the enable
to true. Count, position-cmd, position-fb, velocity-cmd and velocity-fb all
remain at 0.
What next? Start adding the hal commands
Sebastian Kuzminsky wrote:
> [snip]
>
>I totally agree it's far from ideal, but it's the least worst thing I
>could think of at the time.
>
>The fundamental problem is the low-level board drivers need to support
>multiple boards, and each board has multiple pwmgen and stepgen
>instances etc, ea
John Kasunich wrote:
> Eric H. Johnson wrote:
>
>> halcmd loadrt hm2_5i20 config="firmware=hm2/5i20/SVST8_4.BIT num_pwmgens=1"
...
> The kernel module is expecting a single string parameter called
> "config", and you are trying to set its value to
> "firmware=hm2/5i20/SVST8_4.BIT num_pwmgens=1"
>
Eric H. Johnson wrote:
>John,
>
>Thanks, that got it. Does a different parser get called for the hal file?
>Something other than bash? Because that exact syntax does work when run from
>the hal file.
>
>
In general, each bash shell will strip one matching pair of " or '
characters. If you're q
John,
Thanks, that got it. Does a different parser get called for the hal file?
Something other than bash? Because that exact syntax does work when run from
the hal file.
Regards,
Eric
The kernel module is expecting a single string parameter called "config",
and you are trying to set its value
Eric H. Johnson wrote:
> halcmd loadrt hm2_5i20 config="firmware=hm2/5i20/SVST8_4.BIT num_pwmgens=1"
>
>
> => dmesg error <=
> hm2_5i20: Unknown parameter 'num_pwmgens"
>
The kernel module is expecting a single string parameter called
"config", and you are trying to set its value to
"firmware=h
Sebastian,
Here is what I am getting, keystroke by keystroke. I have the symlink in. I
found a broken link which I had erroneously created previously, but I
deleted it and recreated it correctly. It looks to be good when I follow it
from the /lib/firmware folder.
Then, from the terminal session
Stephen Wille Padnos wrote:
> Sebastian Kuzminsky wrote:
>> I agree what's there now is a bit silly... I considered doing it like
>> you suggest, but decided against it. The reason why is that the config
>> info get told to the llio drivers but handled (exclusively) by the
>> generic hostmot2
Sebastian Kuzminsky wrote:
>[snip]
>
>The num_stepgens value gets validated in hm2_stepgen_parse_md().
>
>I agree that the expected integer is not properly validated, and that
>that's not the problem that Eric's seeing, but that it should be done
>better somehow. :-)
>
>
So, do you mind if I
Eric H. Johnson wrote:
> Steve,
>
> Good suggestions. I will try that at my next opportunity. I do not believe
> it to be an issue with spacing or delimiter characters, because I have
> copied the halcmd exactly as it exists in the working .hal file (adding
> halcmd of course). I have tried fully
Steve,
Good suggestions. I will try that at my next opportunity. I do not believe
it to be an issue with spacing or delimiter characters, because I have
copied the halcmd exactly as it exists in the working .hal file (adding
halcmd of course). I have tried fully qualified paths and symlinks and ke
14 matches
Mail list logo
