LinuxCNC 2.8.1 configuration doesn't run after upgrading from 2.7.15

More
5 years 8 months ago #199396 by andypugh
What OS are you running on?

I tried it here on a couple of Debian installs before posting.

It is lower-case-L

Please Log in or Create an account to join the conversation.

More
5 years 8 months ago #199397 by gaston48
ulimit -l (locked memory)

65536

Please Log in or Create an account to join the conversation.

More
5 years 8 months ago #199399 by andypugh
It doesn't seem to be that, then.

If the rtapi logging level is set to 4 or 5 then the shmalloc_dn function lists the free memory and we can see if it is really running out. However setting the rtapi logging level is more difficult than it used to be. I have a very simple HAL component that calls rtapi_set_msg_level(5) when loaded.

(the error is raised here: github.com/LinuxCNC/linuxcnc/blob/master/src/hal/hal_lib.c#L703
which calls: github.com/LinuxCNC/linuxcnc/blob/master.../hal/hal_lib.c#L3020
which calls: github.com/LinuxCNC/linuxcnc/blob/master.../hal/hal_lib.c#L2959

Please Log in or Create an account to join the conversation.

More
5 years 8 months ago - 5 years 8 months ago #199400 by gaston48
If you load my xml file into a config of your own, do you have the same blockage?

I just tried No, on a simple configuration, my xml file works fine ?
Last edit: 5 years 8 months ago by gaston48.

Please Log in or Create an account to join the conversation.

More
5 years 8 months ago #199466 by chris@cnc
I loaded your xml and it also works for me in the basic configuration on my simulator. I remember such problems at the beginning with pyvcp. I had changed the order of the hal files and then it worked. try to keep the basic order of the hal files.


Please Log in or Create an account to join the conversation.

More
5 years 8 months ago - 5 years 8 months ago #199488 by andypugh
I added your PyVCP to sim-axis/axis-9axis
I then loaded a 5i25 and a set of Mesa cards.
All was well.
I then used mux_generic to add a _lot_ of float-type HAL pins and found that the system died at 878 pins.
(actually crashed the PC, but I am running an experimental RTAI kernel)

Trying again I managed to get to 2789 HAL pins before it looked like running out of memory was likely.

Code:
Feb 20 00:04:14 rm-one kernel: [ 642.439483] smalloc_up: shmem available 347376 Feb 19 23:55:10 rm-one kernel: [ 642.439484] HAL: creating pin 'or2.0.in0' Feb 19 23:55:10 rm-one kernel: [ 642.439485] smalloc_dn: shmem available 347280 . . . Feb 19 23:57:12 rm-one kernel: [ 220.570673] smalloc_dn: shmem available 1120 Feb 19 23:57:12 rm-one kernel: [ 220.570677] HAL: creating parameter 'and2.9.tmax' Feb 19 23:57:12 rm-one kernel: [ 220.570677] smalloc_dn: shmem available 1040 Feb 19 23:57:12 rm-one kernel: [ 220.570679] HAL: creating parameter 'and2.9.tmax-increased' Feb 19 23:57:12 rm-one kernel: [ 220.570680] smalloc_dn: shmem available 960

To find the number of pins in your config, open a terminal (with your config running) and "halcmd show pin > pins.txt" then open the pins.txt file in a text editor with line numbers (such as Geany)
Last edit: 5 years 8 months ago by andypugh.

Please Log in or Create an account to join the conversation.

Time to create page: 0.390 seconds
Powered by Kunena Forum