Advanced Search

Ran into an issue with IRQ balance after updating to trixie

Nov 16 19:34:22 hazelnut irqbalance[597]: IRQ 9 affinity is now unmanaged
Nov 16 19:34:22 hazelnut irqbalance[597]: Cannot change IRQ 14 affinity: Read-only file system
Nov 16 19:34:22 hazelnut irqbalance[597]: IRQ 14 affinity is now unmanaged
Nov 16 19:34:42 hazelnut irqbalance[597]: IRQ 123: Override ban to true
Nov 16 19:34:42 hazelnut irqbalance[1572]: /etc/linuxcnc/lcnc_irqbalancepolicy.sh: line 17: /proc/irq/123/smp_affinity: Read-only file system
Nov 16 19:37:52 hazelnut irqbalance[597]: Cannot change IRQ 135 affinity: Read-only file system
Nov 16 19:37:52 hazelnut irqbalance[597]: IRQ 135 affinity is now unmanaged



Turns out there was a bug version of irqbalance that is fixed with

github.com/Irqbalance/irqbalance/issues/308#issue-2217868587

after adding

/etc/systemd/system/irqbalance.service.d/override.conf

➜  ~ sudo cat /etc/systemd/system/irqbalance.service.d/override.conf
[Service]
ProtectKernelTunables=no

It fixed the ethernet mesa issue that was causing latency spikes





 

Sure that's fine, it's basically using widget rules in your Python custom file which should be fine. Just adding a conditional for the button state. If that works roll with it! I don't see an issue with that.

You're loading not in the main hal file but with only one name (instance), you need to add the complete list of names (each instance it is used) and loadrt not should only be in one hal file.

General configuration information:

BoardName : MESA5I25
FPGA Size: 9 KGates5
FPGA Pins: 144
Number of IO Ports: 2
Width of one I/O port: 17
Clock Low frequency: 33.3333 MHz
Clock High frequency: 200.0000 MHz
IDROM Type: 3
Instance Stride 0: 4
Instance Stride 1: 64
Register Stride 0: 256
Register Stride 1: 256

Modules in configuration:

Module: WatchDog
There are 1 of WatchDog in configuration
Version: 0
Registers: 3
BaseAddress: 0C00
ClockFrequency: 33.333 MHz
Register Stride: 256 bytes
Instance Stride: 4 bytes

Module: IOPort
There are 2 of IOPort in configuration
Version: 0
Registers: 5
BaseAddress: 1000
ClockFrequency: 33.333 MHz
Register Stride: 256 bytes
Instance Stride: 4 bytes

Module: QCount
There are 2 of QCount in configuration
Version: 2
Registers: 5
BaseAddress: 3000
ClockFrequency: 33.333 MHz
Register Stride: 256 bytes
Instance Stride: 4 bytes

Module: SSerial
There are 1 of SSerial in configuration
Version: 0
Registers: 6
BaseAddress: 5B00
ClockFrequency: 33.333 MHz
Register Stride: 256 bytes
Instance Stride: 64 bytes

Module: StepGen
There are 10 of StepGen in configuration
Version: 2
Registers: 10
BaseAddress: 2000
ClockFrequency: 33.333 MHz
Register Stride: 256 bytes
Instance Stride: 4 bytes

Module: LED
There are 1 of LED in configuration
Version: 0
Registers: 1
BaseAddress: 0200
ClockFrequency: 33.333 MHz
Register Stride: 256 bytes
Instance Stride: 4 bytes

Configuration pin-out:

IO Connections for P3
DB25 pin# I/O Pri. func Sec. func Chan Sec. Pin func Sec. Pin Dir

1 0 IOPort StepGen 0 Dir/Table2 (Out)
14 1 IOPort StepGen 0 Step/Table1 (Out)
2 2 IOPort StepGen 1 Dir/Table2 (Out)
15 3 IOPort StepGen 1 Step/Table1 (Out)
3 4 IOPort StepGen 2 Dir/Table2 (Out)
16 5 IOPort StepGen 2 Step/Table1 (Out)
4 6 IOPort StepGen 3 Dir/Table2 (Out)
17 7 IOPort StepGen 3 Step/Table1 (Out)
5 8 IOPort StepGen 4 Dir/Table2 (Out)
6 9 IOPort StepGen 4 Step/Table1 (Out)
7 10 IOPort SSerial 0 TXData0 (Out)
8 11 IOPort SSerial 0 RXData0 (In)
9 12 IOPort SSerial 0 TXData1 (Out)
10 13 IOPort SSerial 0 RXData1 (In)
11 14 IOPort QCount 0 Quad-IDX (In)
12 15 IOPort QCount 0 Quad-B (In)
13 16 IOPort QCount 0 Quad-A (In)

IO Connections for P2
DB25 pin# I/O Pri. func Sec. func Chan Sec. Pin func Sec. Pin Dir

1 17 IOPort StepGen 5 Dir/Table2 (Out)
14 18 IOPort StepGen 5 Step/Table1 (Out)
2 19 IOPort StepGen 6 Dir/Table2 (Out)
15 20 IOPort StepGen 6 Step/Table1 (Out)
3 21 IOPort StepGen 7 Dir/Table2 (Out)
16 22 IOPort StepGen 7 Step/Table1 (Out)
4 23 IOPort StepGen 8 Dir/Table2 (Out)
17 24 IOPort StepGen 8 Step/Table1 (Out)
5 25 IOPort StepGen 9 Dir/Table2 (Out)
6 26 IOPort StepGen 9 Step/Table1 (Out)
7 27 IOPort SSerial 0 TXData2 (Out)
8 28 IOPort SSerial 0 RXData2 (In)
9 29 IOPort SSerial 0 TXData3 (Out)
10 30 IOPort SSerial 0 RXData3 (In)
11 31 IOPort QCount 1 Quad-IDX (In)
12 32 IOPort QCount 1 Quad-B (In)
13 33 IOPort QCount 1 Quad-A


When using:

setp hm2_5i25.0.stepgen.03.direction.invert_output
setp hm2_5i25.0.stepgen.03.step.invert_output

Results:

Debug file information:
Note: Using POSIX realtime
./Autumn.hal:240: setp requires 2 arguments, 1 given
12304
Stopping realtime threads
Unloading hal components
RTAPI_PCI: Unmapped 65536 bytes at 0x7f370745c000
Note: Using POSIX realtime


But:

setp hm2_5i25.0.stepgen.03.direction.invert_output TRUE
setp hm2_5i25.0.stepgen.03.step.invert_output TRUE

Executes and launches

I read a few posts that you said on this but it failed to compile when launching linuxcnc when I used that stepgen/invert nomenclature. Past posts included 1 or TRUE in the definition, I will try the above verbatim.

Sorry for the delay I'm in the process of swapping dev PC's around so I can test on a clean install. In the interim. Check and see if your tool table has pockets listed in the file. If so, make sure the pocket and roll numbers are the same. Something in latest release of linuxcnc changed with pockets that I think is making a bit of a mess in some areas.

Yes, you have a bit of a mess. chat gpt did you no favors. just follow the instructions. And first and foremost, verify the probe basic SIM works correctly. That will let you know if the installation is correct as it has correct configs.

Once verified the SIM works, take your time and go line by line through the doc and you will have a machine at the end. Start with a good working axis configuration..

Linux OS & kernel type
Debian GNU/Linux 12 (bookworm)
6.1.0-30-rt-amd64 #1 SMP PREEMPT_RT Debian 6.1.124-1 (2025-01-12) x86_64 GNU/Linux

Version of Linuxcnc
2.9.6

Version of QtPyVCP and Probe Basic
0.0 & 2.9.6

Howdy folks,
Forgive me in advance as this project is the first time I've touched linux, linuxcnc, probe basic, or attempted a CNC conversion. I used the quickstart guide to install, and I've gone through the hal and ini files. Although, both the hal and ini files could be severely incorrect now. I believe I have the necessary config files.

Now, admittedly, I've used ChatGPT way way too much for this, so I might need to start from scratch.

So, after installing linuxcnc, the sim.axis worked great, X and X moved well and were moving roughly correct distances (I don't have my Z installed yet).

I then installed probe basic and went through pncconfig to make my custom config file that opens into probe basic. In that custom config I can't get anything to work. The E-stop button freezes the screen for a few seconds, but it never disables the E-stop.

And now when I open into sim.axis, everything works but my motors don't actually move. E-stop disables, machine turns on, and jogging shows in the graphics.

Just trying to get some motors to move and maybe make a circle with gcode. That'll be a step in the right direction. HAL & INI are attached. Also I'm running a Lenovo M710Q with a Mesa 7i96s.

Thanks in advance

 
 

 

On the 6I25/5I25 mesaflash lists DB25 pin numbers

DB25 connectors are numbered:

1    2    3    4    5    6    7    8    9    10    11    12    13
  14  15  16  17  18 19  20  21  22    23     24    25

Also you don't need to use GPIO to invert step or dir pins, there are aliased names like:

hm2_5i25.0.stepgen.00.direction.invert_output

hm2_5i25.0.stepgen.00.step.invert_output



 

I believe it's Bob if I did the flash right. I'm still not 100% on how all these talk and write to one another. I'll run the readhmid after I am done making dinner.

I'm not sure if this nomenclature starts at 0 or 1. It's physically the 8th pin starting from 1 which is how I understand DB25 to be counted by, but the config starts joints from 0 and I'm not sure how to describe the count due to that.

2.9.7, Preempt rt ISO

Stepper online, no clones.

axis

The drivers, CLXX's no daughters.

I think the problem being between the 6i25 and the drivers is probable.

I tried some weak oscilloscopes. Nothing better than AC multimeters, it just read gibberish. I really should get a good one as a work expense. Something that can ID freq bands. I could go that route if this SHOULD work.

So, to be more coherent on the control desires....

1: A new UI skinning of (probably) qtdragon, possibly vertical layout. I'll do the layouts, I'll need help turning it into a real working UI.

2: A linuxcnc build with a configuration wizard that includes ethercat and modbus, atc types, and some specific presets for lemontart (and then presets for any other machine people like - maybe presets can be just saved out from peoples manual configurations and shared).

3: Jerk limiting.at minimum on any exact top moves. Secondary is to add it to other motion. Bonus points for integrating all the fancier TP stuff from the tormach 1500.

For the physical control, I'm going to do a printed panel, with waveshare touch display, and a common low cost passively cooled dual nic mini itx board. Just so its easy and standard. "go buy these things, assemble, load linuxcnc onto it and run set up wizard"

What firmware are you using?

sudo mesaflash --device 5i25 --readhmid

Will print the pinout of the currently installed firmware

( assuming you are using the 5abob firmware, DB25 pin 7 is stepgen 2 DIR, not STEP )
 

And we're done. Coordinating to start production tonight. Eep! (goes faster when someone else does all the work... and pays for it. )

What version of LinuxCNC are you using?

Did you install from the ISO?

Did you buy those drivers from stepperonline? (there are clones of the clones)

What GUI are you using (I assume AXIS)?

What does the 6i25 connect to?

Perhaps the problem is whatever is between the 6i25 and the drivers.

The pulse pin carries high frequancy pulses, don't trust a multimeter to reliably measure it, especially when it's set to DC.

Try setting your multimeter to AC, or use an LED+current-limiting resistor to view the pulses.