7i84 pin does not exist

A little confusing --
In the main machine.hal file..

The config IS working with the first of the 7i84 boards - hm2_7i93.0.7i84.0.0

But with the second - hm2_7i93.0.7i84.0.1

If I reference
hm2_7i93.0.7i84.0.1.output-00-invert
I get as an command line erorr :
Pin 'hm2_7i93.0.7i84.0.1.output-00-invert' does not exist

Yet - if I start up linuxcnc without the offending line

I can:
# halcmd show all |grep hm2_7i93.0.7i84.0.1.output-00-invert
26 bit RW FALSE hm2_7i93.0.7i84.0.1.output-00-invert

So the pin really does exist - and I'm using a similar pin on the first board without any trouble?

( I can also see the pin with halmeter and set it true with halcmd.. )..

,.,.,.
These boards are loaded with :

loadrt hostmot2
loadrt hm2_eth board_ip="192.168.10.10" config=" num_encoders=4 num_pwmgens=4 num_stepgens=4 sserial_port_0=00000000"

typo or editor issue maybe?
You might try accessing the pin via halcmd
(which has command completion)

I checked the file for non ascii characters..

I WAS able to access the pin via halcmd.. I can see it - read it - write to it.

Wondering if it takes time for the second boards to be ready?

It works if I comment it out of the machine.hal file..

Is there a way to put a delay in the hal file?

No, as soon as LinuxCNC has started, all pins will be available

This works
But this does not
The second set returns -
./wells.hal:316: Pin 'hm2_7i93.0.7i84.0.1.output-00-invert' does not exist

I'm thinking the error is blaming the wrong pin? This does not make sense.

I've tried linuxcnc -d - I see

And later

That's correct. sorry i missed it on the first reply
There is no such pin as hm2_7i93.0.7i84.0.1.output-00-invert

(its a parameter that conditionally inverts the corresponding output pin)

OK - looks just like a pin from
$ halcmd show all

I should have used
$ halcmd show pin

And when I set it true - it toggled the output fooling me ..
Tested and working now.

$ halcmd setp

Sets both pins and parameters..

I'm using iocontrol.0.user-enable-out it control a relay in the physical estop loop.

Thanks for you patience and help..

,.,.
I dug through the source code a bit looking at etop - you might know if there is any internal function that detects software fail - like an old style watchdog - and hits estop?

The halui.estop pins don't seem to be connected to anything I can find and I didn't find anything suggesting that the iocontrol pins associated with estop reflect anything internal.

(I think there may be an historical reason for iocontrol vs halui ?)

Not built in but there is a watchdog hal component

halui pins are connect to linuxcnc through NML messages.
If linuxcnc send an ESTOP NML message, halui's pins will reflect this.

IOcontrol does the same thing, it lets an integrator tap into the estop chain.

There are many reasons linuxcnc sets the estop - limit switches, following errors etc.

@cmorley
I think I get it now - i missed that the halui.estop inputs are edge activated. (not in the docs).

iocontrol.0.user-enable-out is the inverse of halui.estop.is-activated

halui.estop.activate on rising edge activates halui.estop.is-activated
halui.estop.reset on rising edge resets halui.estop.is-activated

Levels won't prevent one from overriding with the axis button.
,.,.
I found that iocontrol.0.user-request-enable outputs a pulse - which might have been for connect to the halui.estop input signal..