#linuxcnc-devel Logs

Mar 25 2023

#linuxcnc-devel Calendar

09:05 AM andypugh: pcw--home: Your mesaflash script works, and a modified version reads data from my USB/Serial thing. So it’s looking like problems inside my hm2 code.
09:39 AM JT-Cave: tMax = readtmax + writetmax and packetTime = tMax / cpuSpeed there was some scaling on cpuSpeed I think but I've lost my notes
09:39 AM JT-Cave: scaling depending on if it's MHz or GHz iirc
09:52 AM pcw--home: The (X86) tmax values are in CPU clocks so to get the percentage of servo thread time these represent you must divide them by CPU clocks per full servo period = servo thread period in seconds*CPU clock speed in Hz
10:03 AM JT-Cave: thanks
10:17 AM andypugh: pcw--home: I am not having much luck sending multiple packets with the mesaflash script.
10:17 AM andypugh: https://paste.debian.net/1275236/
10:18 AM andypugh: Write 16 bytes to 6100 then write 8 twice to 6200
10:22 AM andypugh: But, I am also wondering if that is important, as we probably want to send a packet, then wait for the response, rather than try to write a second command over the response ?
10:26 AM pcw--home: http://freeby.mesanet.com/2mf (2 packet sequence)
10:27 AM pcw--home: but yes, modbus is full synchronous so never any need to deal with more than 1 packet send or receive
10:36 AM andypugh: Hmm, my 2-packet script looks much like yours (I wasn’t worrying about reading the data, just looking to see two packets were reported as recieved)
10:42 AM pcw--home: This is what I get:
10:42 AM pcw--home: 3FE214C8
10:42 AM pcw--home: 00000008
10:42 AM pcw--home: 54535251
10:42 AM pcw--home: 58575655
10:42 AM pcw--home: 3FE114C8
10:42 AM pcw--home: 00000008
10:42 AM pcw--home: 64636261
10:42 AM pcw--home: 68676665
10:43 AM pcw--home: 2 packets, both 8 bytes
10:44 AM Tom_L: the 3F line the start/stop ?
10:45 AM pcw--home: that line is the RX mode/status register, the first one shows 2 RX packets
10:46 AM pcw--home: the 8 on the next line is the number of chars in the first packet
10:46 AM andypugh: I need to think what I was testing. I think I have TX mask turned on so it’s not a loop-back. But I wasn’t seeing two packets on the scope
11:44 AM andypugh: Yeah, your script works in loop-back.
11:46 AM andypugh: pcw--home: Another thing, I have here a 6i25 that I offered to test for a forum user. His PCs don’t boot with it. When I take my own 6i25 out and put his in, mine doesn’t boot either. So it’s pretty clearly a card fault. Any easy checks and easy fixes?
11:49 AM pcw--home: make sure its jumpered for 33 MHz.
11:51 AM pcw--home: Also possibly short put the EEPROM data pin at power up so it cannot configure. If the PC boots, its like been programmed with a wrong bitfile
12:02 PM linuxcnc-build2: Worker `checkin` is missing. It was seen last on Sat Mar 25 10:01:40 2023.
03:23 PM linuxcnc-build2_: Worker `checkin` is missing. It was seen last on Sat Mar 25 13:22:02 2023.
04:08 PM linuxcnc-build2_: Build [#594](http://buildbot2.highlab.com/buildbot/#builders/8/builds/594) of `30-dsc-uspace.debian-12-bookworm` 3completed successfully.
04:14 PM linuxcnc-build2_: Build [#24](http://buildbot2.highlab.com/buildbot/#builders/20/builds/24) of `50-deb-uspace.debian-08-jessie-i386` 4failed.
04:24 PM linuxcnc-build2_: Build [#692](http://buildbot2.highlab.com/buildbot/#builders/13/builds/692) of `00-checkin` 3completed successfully.
05:05 PM linuxcnc-build2_: Build [#595](http://buildbot2.highlab.com/buildbot/#builders/8/builds/595) of `30-dsc-uspace.debian-12-bookworm` 4failed.
05:13 PM linuxcnc-build2_: Build [#693](http://buildbot2.highlab.com/buildbot/#builders/13/builds/693) of `00-checkin` 4failed.