Linux will not recognize my BOB no matter WHAT I try to do

Again..... My ENTIRE frustration.......... This SHOULD BE WORKING. It is SUPPOSED TO be working.... I KNOW the MosChip 9900 driver will work. The computer can CLEARLY see it. I just cannot get the computer to TALK with it REGARDLESS of what I type into the HAL file. Entering cfg="0xe010 out" yields the exact same results as everything else I have tried thus far. I can start the Axis GUI and the relay on the BOB powers low, the GUI allows me to "jog" the machine virtually (although it obviously doesn't allow the motors to move), and I still get the exact same "Linux parallel port @57360 not found" error that is pictured in the last post picture of the GUI screen.

To again bring light back to what I believe might be the source of the issue (even though I am admittedly still a rookie with Python and Linux), I had mentioned earlier that if you look in the cat /proc/ioports output text file that I attached a few posts ago, you can see that the parallel port is the ONLY thing that does not have a kernel driver running any module of any kind. EVERY SINGLE example that I have come across in ANY online searching across ANY platform shows the parport_pc kernel driving that specific module. If there is no kernel module driving the parallel port, then that could explain why the parallel port can't be talked to, right? So changing the address of where the computer is trying to send the signals to will do nothing if the parallel port the computer is trying to send the signals to isn't even being engaged by the kernel, are my assumptions correct? If they are, then I would also assume that the issue would then lie in initializing the kernel driver, which I cannot do. I cannot insmod the module. I know where it is in the file system, but no terminal commands I have been able to find or videos I have watched on doing that will work. Perhaps I am syntaxtually challenged as well as synaptically challenged, but I have tried every terminal command I have come across and I still have no parport_pc kernel module driving my parallel port....
Go to $ /lib/modules/4.19.0-11-amt64/kernel/drivers/parport in Buster 2.8.0 and it is sitting right there. The file is called parport_pc.ko. I just can't get it into the programming where I believe it needs to be to initialize the parport_pc kernel driver upon boot up. Does that sound like a plausible source of the issue to anyone else?

Really not trying to sound like I know what I am talking about Like I said, I am admittedly still a rookie to this system. I am just trying to reinject a possible different thought path to the solution. I've spent, on average, literally about 18 hours per day for the last 2 + weeks trying to figure this problem out because I need this machine to be operational for my business SO badly. This is the most pressing issue on my plate at the moment, so I am just trying to help solve the problem by bringing up what I have found in my troubleshooting may be part (if not all ) of the issue (maybe?).

Kidkasual1 wrote: I had figured out that I had a WCH 382 chipset in the parallel port adapter I had purchased, which, again..... see prior posts... (what KILLS me is that it was actually WORKING before it didn't... How is the WCH 382 chipset KNOWN to not work with Linux, yet I honestly had the machine jogging on command in all 4 axis before it all crashed........HOW??????)

This is a dual-boot system and these boards need a driver loading in to them at power-up. Perhaps booting in to Windows loaded the driver to the card. Then, booting to Linux without a power-down would leave the WCH board on a working state.

If you were then to power down the system and boot straight to LinuxCNC, the driver would not be resident on the board, and the port would stop working.

Now that you have reinstalled Windows you probably don't have the WCH driver anywhere on the system.

Kidkasual1 wrote: If there is no kernel module driving the parallel port, then that could explain why the parallel port can't be talked to, right?

That isn't a problem, as LinuxCNC provides its own driver.

However, the question of _why_ the parport_pc driver isn't trying to drive the card is possibly indicative of a problem.

But, you say that you hear the relay click on the board when you power up LinuxCNC? That actually hints that there is some form of communication.

You also say that the stepper motors are not locked. This rather suggests that the issue might be with the drive enable wiring. I would suggest just disconnecting the drive enable (As most steppers default to "enabled") and see if the drives do, at least, then lock.

Reading the docs at :
linuxcnc.org/docs/html/man/man1/hal_parport.1.html

It looks like the correct syntax might be :

loadrt hal_parport cfg="0xe010 out"

And, of course, also try 0xe000

Jon

OHHH! Some BOBs need more drive current on the data lines that the typical parport can supply. The trick is to set the parport to EPP mode in the loadrt line.
Instead of setting it to "out", set it to "epp" and see if that works.

Jon

Thank you all for you replies and input. As for the dual boot, I only have Windows on the hard drive because I had single booted Linux with the first install, and when my computer crashed and I wasn't able to reclaim it, it spooked me a little bit. As for now, I really only boot straight into Debain upon power-up. I don't really even boot Windows on that computer unless I actually have to for some reason. It's really just there for peace of mind.

I also changed from the X-Media PCIe adapter card that runs with the WCH 382X chipset and driver to the MosChip 9900 chipset and driver that runs on a new PCIe card from SIIG about a week ago. Same results. Nothing.

Also rewired everything 3 separate times to 3 separate BOB'S from 2 different manufacturers. Same results. No communication.

Tried setting parport to epp and also the e000 address with no results.

Set pin 14 (enable) to "unused", and nothing.

As for the relay, it powers high upon boot up and stays high. It only powers low when I try to open the axis GUI, and even then only if the parallel port address is one it recognizes. e010 work fine for that. It will stay low until I power down the entire computer. However, when I start the Axis GUI, it still gives me the same "Linux parallel port @57360 not found" error in the bottom right corner of the GUI display and I have ZERO communication with the BOB.

As Andy said, If LinuxCNC changes the relay state, it has access to the port
so that really suggests a hardware problem. If the step motors are not locked
(and you don't have them disabled somehow). That is your issue, not LinuxCNC
setup.

What REALLY REALLY confuses me is that when I installed Linux the very first time using this EXACT same hardware, with this EXACT same wiring, in this EXACT same configuration, I had FULL control of the machine. I was able to jog and control all 4 axis. ACTUALLY.... I was running the X-Media PCIe card with the WCH382 chipset in it - that isn't supposed to work - and it ALL WORKED. AMAZINGLY well. Then it just all went to crap. That is why I am SO freaking perplexed. It ALL WORKED until it DIDN'T. And now I can't get it back to working order and I have ZERO clue as to why, other than I can see that the parallel port is not initiating the parport_pc kernel.......