Problems with building Linux CNC source

Robert, you could have a package version of 2.8 and a run in place version of master on the same machine or even multiple run in place versions of different types.

thanks Phil, 

I did a RIP years ago .... but have got lazy.

I'll have to try again with a RIP for my mill. lathe and router, and leave the buildbot for QTPlasmaC (which has been great for up to date now it's beta and just being honed)

This document :
-> linuxcnc.org/docs/2.8/html/code/building-linuxcnc.html
Is a little bit misleading. It should be explicitly mentioned that this command:
-> git clone git://github.com/linuxcnc/linuxcnc.git linuxcnc-dev
get you the master copy which is development version. For production version download 2.8 software package.
I have downloaded the code under assumption that is 2.8.2 as in official package but it was not.

smal wrote:

The above mentioned link:
-> linuxcnc.org/docs/2.8/html/code/building-linuxcnc.html
does not work with 2.8.2 because using this command:
-> git clone git://github.com/linuxcnc/linuxcnc.git linuxcnc-dev you get the newest 2.9.0 master version of the source code.
I have made it exactly as described in this document. The results are the problems which I am facing now.

if you are in the ~/linuxcnc-dev/src folder you should be able to get to 2.8.2 by
git checkout 2.8
git pull
make clean
make
sudo make setuid

There are many other branches in Linuxcnc which appear in a drop down in git
You may need to resolve some dependencies

smal wrote:

This document :
-> linuxcnc.org/docs/2.8/html/code/building-linuxcnc.html
Is a little bit misleading. It should be explicitly mentioned that this command:
-> git clone git://github.com/linuxcnc/linuxcnc.git linuxcnc-dev
get you the master copy which is development version. For production version download 2.8 software package.
I have downloaded the code under assumption that is 2.8.2 as in official package but it was not.

 

to clone into 2.8 try:

I have copied directly 2.8.2 zip file to my folder. Then I have unpacked it and everything worked as expected till now. So the problem with compiling and executing application is solved.
Again thank you very much to all of you who have invested some effort and time to achieve this state.

Can linuxcnc be run on more than one core? Or is this valid only for dependencies or compiling? Actually when compiling 2.8.2 software package there was no need for additional download to resolve dependencies problem.
This question now is a little bit offtopic:
I am running pcie Mesa 6i25 card. Under assumption that everything what is important runs in a fpga is realtime performance of the linuxcnc really needed?

running multiple cores is just for "make" to speed up the build process

you still need to be running a RT (real time) patched OS.... if you installed from linuxcnc iso / usb stick you will be fine, if you went another route (mint / MX-linux) then you may need to patch the os.

there are two loops.... the "base loop" and the "servo loop".

using mesa hardware the heavy lifting (base loop) is done on the FPGA at a much higher loop speed.

the servo loop (1khz) is still required to send data from the planner to the Mesa fpga and get information back from the fpga.

Here are the results of the measurement:
I have compiled two fresh software linuxcnc packages one with option -j2 and one without. I am using an ancient processor on my test platform. Command lscpu gives me following:

Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
Address sizes: 36 bits physical, 48 bits virtual
CPU(s): 4
On-line CPU(s) list: 0-3
Thread(s) per core: 2
Core(s) per socket: 2
Socket(s): 1
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 42
Model name: Intel(R) Core(TM) i3-2100 CPU @ 3.10GHz
Stepping: 7
CPU MHz: 1698.108
CPU max MHz: 3100.0000
CPU min MHz: 1600.0000
BogoMIPS: 6185.89
Virtualization: VT-x
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 3072K
NUMA node0 CPU(s): 0-3
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 popcnt tsc_deadline_timer xsave avx lahf_lm epb pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm arat pln pts md_clear flush_l1d

According to datasheet it is a 2 core processor.
So compiling with option -j2 gives me 2 minutes and 23 seconds compile time
Without this option the compile time is 3 minutes and 53 seconds.
Conclusion: option - jX works very well as the difference is substantial.