New litexcnc connections="eth:...." problem #38
-
|
I have Problems with the "new" loadrt standard. The new single line loading produces read / connection errors. Now when I try to install driver after pulling the latest "11" I get this... INFO: Compiling LitexCNC driver... |
Beta Was this translation helpful? Give feedback.
Replies: 16 comments 10 replies
-
|
https://github.com/Peter-van-Tol/LiteX-CNC/files/12503774/Semse.hal.txt |
Beta Was this translation helpful? Give feedback.
-
|
Brilliant! I got a merge conflict when releasing version 1.0.0. This merge conflict was resolved, however not correctly. When testing the release, the test took on the old driver-files, causing the error not to be caught before release. I'm promoting this to an issue, which has to be resolved ASAP. |
Beta Was this translation helpful? Give feedback.
-
|
Version 1.0.1 has just been released, solving the error. Can you please try: pip install -U litexcnc |
Beta Was this translation helpful? Give feedback.
-
|
Good morning. Going to test it on my maschine maybe tonight or tommorow |
Beta Was this translation helpful? Give feedback.
-
|
Any information about my error in the hal? |
Beta Was this translation helpful? Give feedback.
-
|
Not yet, will check tonight. Might be resolved with version 1.0.1 al together, because the error might stem from a previous patch.. |
Beta Was this translation helpful? Give feedback.
-
|
The HAL seems correct to me, could you please try with v1.0. to test my theory that it might have been a previous patch. If the problem persists, could you please post here the complete output from Linuxcnc. I find it easy to start LinuxCNC from the command-line and pipe the output to a file, i.e.: linuxcnc > ~\output.txt |
Beta Was this translation helpful? Give feedback.
-
|
Same problem with connections="eth:...." in halrun the connections="eth:...." starts with no problems... |
Beta Was this translation helpful? Give feedback.
-
|
here is when I type it one by one in halrun |
Beta Was this translation helpful? Give feedback.
-
|
Ok...forget to enter "start" in halcmd....then the show begins |
Beta Was this translation helpful? Give feedback.
-
|
When the component is started the communication is OK, because the card is initialized and there is communication with the card. However, as soon as data is being written and read from the device. It is strange that after initialization while in the loop the card crashes and does not respond. I cannot reproduce with your the files I have until now. Could you please share the following:
|
Beta Was this translation helpful? Give feedback.
-
|
this is the version that I use. in a terminal I change to the Litex folder....then
|
Beta Was this translation helpful? Give feedback.
-
|
I've tried installing the driver from your zip-drive, and I cannot reproduce the error. Next week I'll get a new 5A-75E in (last one is now in a machine) and will flash that with your firmware, based from the dropbox. Would it be a possibility to set-up a digital meeting somewhere to investigate together? |
Beta Was this translation helpful? Give feedback.
-
|
Great, which timezone are you in? |
Beta Was this translation helpful? Give feedback.
-
|
Germany, Kleve, near dutch border. |
Beta Was this translation helpful? Give feedback.
-
|
That's the same timezone, which makes it easier. I'm 'just' across the border located in Rotterdam. Would Thursday suit you, preferably around 21:00? |
Beta Was this translation helpful? Give feedback.
-
|
Should be possible Is this your email? ? |
Beta Was this translation helpful? Give feedback.
-
|
That one is correct! |
Beta Was this translation helpful? Give feedback.
-
|
Beta Was this translation helpful? Give feedback.
-
|
To prevent this error from happening, a temporary fix can be implemented which sets the header on the data every time, at the cost of some efficiency. This does not solve the cause of the problem though ... @OJthe123 : Tonight I'll prepare a version of the driver which prints out the address buffer in each step. Would you be willing to test this to find which buffer is misbehaving. |
Beta Was this translation helpful? Give feedback.
-
|
Sure. |
Beta Was this translation helpful? Give feedback.
-
|
You can test a new version tonight, because I found the culprit: // Clear buffer (except for the header)
memset(
litexcnc->fpga->read_buffer + litexcnc->fpga->read_header_size,
0,
litexcnc->fpga->read_buffer_size
); In the read function I clear the read buffer. The size of the read buffer is equal to the data + header. When cleraring the buffer, only the data is cleared, therefore in above code the offset The code above should thus be : // Clear buffer (except for the header)
memset(
litexcnc->fpga->read_buffer + litexcnc->fpga->read_header_size,
0,
litexcnc->fpga->read_buffer_size - litexcnc->fpga->read_header_size
); |
Beta Was this translation helpful? Give feedback.
-
|
@OJthe123 : you can pull branch #44 and test the modification. When the read errors disappear, I'll roll this version out. |
Beta Was this translation helpful? Give feedback.
-
|
Yes! |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for testing, will merge and release a new version. Can you mark my reply above as 'Mark as answer'? |
Beta Was this translation helpful? Give feedback.
@OJthe123 : you can pull branch #44 and test the modification. When the read errors disappear, I'll roll this version out.