#linuxcnc Logs

Mar 06 2025

12:17 AM lcnc-relay: <oddname_skane@> rs: how would this be possible?
12:31 AM lcnc-relay: <oddname_skane@> I had a apt get install version, updated to buildbot for reasons I dont know but it fixed the issues I had
12:52 AM Deejay: moin
01:01 AM lcnc-relay: <oddname_skane@> rs: Can i check in an easy way if this is the case with mixed installations?
01:13 AM lcnc-relay: <oddname_skane@> it does seem as if im using 2.10 though
01:13 AM lcnc-relay: :2.10.0~pre0.4165.g55fb1a0267
01:32 AM lcnc-relay: <meisterdippel@> moin
04:30 AM Tom_L: morning
04:32 AM -!- #linuxcnc mode set to +v by ChanServ
04:45 AM lcnc-relay: <oddname_skane@> moin moin
05:20 AM lcnc-relay: <oddname_skane@> or say. if I were to remove linuxcnc completely how can I be sure to get the patched up version that should work?
06:50 AM JT-Cave: the master branch is never stable
09:45 AM lcnc-relay: <rs> oddname_skane@: 2.10 is not stable, but usually it should work without problems. but it is actively developed and there may be the odd issue if you keep updating it constantly. I would recommend monitoring github issues if you run 2.10.
09:46 AM lcnc-relay: <rs> oddname_skane@: something with your remap mixes stuff from 2.10 with stuff that still tries to do it like in 2.9
10:00 AM lcnc-relay: <oddname_skane@> 2.9 worked betterz new error now that is likely due to the plugin 🙂
02:09 PM roycroft: my new office desk chair should be delivered any time now - and not a moment too soon
02:09 PM roycroft: it's become rather painful sitting in my current one for hours at a time
02:10 PM roycroft: i guess the 20 year warranty has expired :)
02:10 PM xxcoder: you sure its not your butt and spine that have warranty expired?
02:11 PM roycroft: that too
02:11 PM roycroft: but the new chair should give them better support
02:13 PM xxcoder: nice
04:08 PM rdtsc-w: Hmm... upgraded 7C80 RPi4B for a 5B with latest Rpi5 .xz from https://linuxcnc.org/downloads/ . It boots, /dev/spidev0.0 exists, but mesaflash can't see the 7c80; always responds with bad cookie message. At power-on, LED CR56, 57, and 58 all come on for maybe 0.1s, then only CR58 is left illuminated - so the 7c80 seems to be working. Ideas?
04:13 PM JT-Shop_: we used to have an office depot now it's fengs
04:17 PM lcnc-relay: <rs> does it say the cookie value?
04:40 PM roycroft: finally, my chair is being delivered
04:40 PM * roycroft can feel the pain going away already, in anticipation of having a comfy chair
04:44 PM lcnc-relay: <lolthekidison@> Hope everyone has been doing well. Is it possible to use one of the parralel port output pins as a pwn signal?
04:45 PM Tom_L: probably
04:45 PM Tom_L: limited to the port speed
04:49 PM Tom_L: http://linuxcnc.org/docs/stable/html/hal/parallel-port.html
05:13 PM rdtsc: rs, cookie value is all zeroes - it doesn't see the card at all
05:16 PM rdtsc: seeing this in dmesg: [ 1.201292] rp1-pio 1f00178000.pio: error -ENOENT: failed to contact RP1 firmware
05:17 PM roycroft: aah, no more sitting on springs
05:19 PM roycroft: does anyone know when the 7i96s might be available again?
05:21 PM Tom_L: pcw has 63 available
05:22 PM Tom_L: https://store.mesanet.com/index.php?route=product/product&product_id=374&search=7i96s
05:25 PM roycroft: hmm, i was having a hard time finding the ordering page on his site yesterday
05:25 PM roycroft: and jt is out of them
05:25 PM Tom_L: store.
05:25 PM JT-Shop_: I'm getting a bunch next week
05:25 PM roycroft: i was having a bunch of browser problems yesterday, though, so i won't blame pcw
05:25 PM roycroft: nice, jt-shop!
05:26 PM roycroft: i'm not sure i am going to order one or two, but i know i have the option now
05:26 PM roycroft: it depends on where i decide to install components
06:13 PM rdtsc: rs, found out using `rpi-eeprom-update' that this pi5 has a firmware from 2024-09-23; updated to 2025-01-22, but still no mesaflash joy.
06:19 PM lcnc-relay: <captainhindsight_.@> https://static01.nyt.com/images/2025/03/06/multimedia/06athena-moon-landing-write-thru-01-bqgp/06athena-moon-landing-write-thru-01-bqgp-superJumbo.jpg space drill
06:19 PM lcnc-relay: <ccatlett1984@> rdtsc, have you tried the image that ended up working for me? , do we have it narrowed down to an issue with the image/build?
06:23 PM lcnc-relay: <captainhindsight_.@> 7i96 $119, 7i96s $149
06:24 PM lcnc-relay: <captainhindsight_.@> why choose the s vs the non-s version?
06:26 PM lcnc-relay: <captainhindsight_.@> S version uses Efnix vs non s version using Xilinx
06:30 PM lcnc-relay: <captainhindsight_.@> are there other differences between the two versions?
06:36 PM Tom_L: rdtsc, what mesaflash version?
06:37 PM Tom_L: capn, s has analog for spindle control
06:37 PM Tom_L: rdtsc, it's up to 3.5.10 now
06:38 PM Tom_L: capn, non s does not
06:42 PM lcnc-relay: <ccatlett1984@> I built from source on my pi5
06:42 PM lcnc-relay: <ccatlett1984@> since .10 hadn't been released yet
06:42 PM Tom_L: i always do
06:43 PM lcnc-relay: <captainhindsight_.@> 7I96S "A high speed encoder interface is provided for spindle synchronized motion and an isolated analog spindle output allows spindle speed control ." I wish that was in bold text or similar.
06:44 PM * Tom_L gives capn a highliter
06:48 PM lcnc-relay: <skunkworks8841@> wow - lost another starship
06:49 PM lcnc-relay: <skunkworks8841@> but stuck another booster landing
07:10 PM Tom_L: skunkworks, it keeps my kid in business making parts :)
07:10 PM lcnc-relay: <skunkworks8841@> sure - I don't think it is a big deal. They just have to figure out why version 2 is failing. Version 1 was pretty darn successful.
07:10 PM rdtsc: tom, just built Mesaflash version 3.5.11
07:11 PM xxcoder: fails means more money for machinists
07:11 PM lcnc-relay: <skunkworks8841@> they do need to start re-using boosters though
07:11 PM Tom_L: wonder what he did between .10 and .11
07:11 PM rdtsc: ccatlett1984 haven't tried that one yet... guess I can find another sd card around somewhere
07:21 PM lcnc-relay: <skunkworks8841@> I still can't believe they are catching the booster as well as they are..
07:21 PM lcnc-relay: <skunkworks8841@> crazy
07:33 PM Tom_L: technology has come a ways since apollo
07:34 PM xxcoder: wish we had a nice non-chemical way to launch stuff. it would lower prices by a whole lot
07:34 PM xxcoder: sky tower is still too hard
07:35 PM lcnc-relay: <skunkworks8841@> well - if spacex can pull this off - space flight will be pretty cheap. (relatively...)
07:35 PM lcnc-relay: <skunkworks8841@> It will be almost the cost of fuel.
07:35 PM Tom_L: out of my price range
07:35 PM lcnc-relay: <skunkworks8841@> me too
07:36 PM lcnc-relay: <skunkworks8841@> I still like the anaolgy.. What if every airplane was destroyed after every flight. Can you imagine how expensive flight would be?
07:36 PM lcnc-relay: that is how it was done before spacex started re-using boosters..
07:37 PM lcnc-relay: <skunkworks8841@> if they can put 100-150 tons into orbit for just fuel cost? that will be amazing.
08:15 PM lcnc-relay: <captainhindsight_.@> nah, service fees, destination fees, reentry fees, 911 fees, excise tax fees, DOGE fees.....
08:21 PM lcnc-relay: <big_kevin420@> rdtsc do you have SPI turned on?
08:21 PM lcnc-relay: <big_kevin420@> and does linuxCNC connect to the mesacard?
08:40 PM rdtsc: of course kevin, /dev/spidev0.0 exists but mesaflash says bad cookie and all zeroes. About to try older Rpi5 image
08:50 PM rdtsc: ccatlett1984, the earlier 6.1.61-rt15 just kernel panics immediately at start of boot
08:53 PM lcnc-relay: <ccatlett1984@> odd
08:54 PM lcnc-relay: <ccatlett1984@> i've reflashed that twice now, with good results both times
08:55 PM lcnc-relay: <ccatlett1984@> this img? rpi-5-debian-bookworm-6.1.61-rt15-arm64-ext4-2023-11-17-1520.img.xz
08:59 PM rdtsc: yep, gonna try flashing it again
09:39 PM lcnc-relay: <big_kevin420@> well, its not assumed you have it turned on, just try to turn it on again, restart, and then try mesa flash
09:39 PM lcnc-relay: <big_kevin420@> but also get with linuxcnc
09:39 PM lcnc-relay: <big_kevin420@> thats really the only thing that matters
09:41 PM rdtsc: tried re-flashing 6.1.61-rt15, now get the non-boot screen (sd card bad image, can't read parition) hmm one more time maybe
09:42 PM rdtsc: wow okay, sd card comes up as completely blank... weird
09:53 PM lcnc-relay: <big_kevin420@> jsut keep trying with your working one
09:54 PM lcnc-relay: <big_kevin420@> enabled using raspi-config or the one menu in the start screen?
09:56 PM rdtsc: on debian 12, enable spi by editing /boot/broadcom/config.txt - raspi-config was replaced with menu-config, which is quite limited
09:58 PM lcnc-relay: <big_kevin420@> right debian
09:58 PM lcnc-relay: <big_kevin420@> uhhh what was the deal with that
09:58 PM lcnc-relay: <big_kevin420@> theres 2 locations for that file when using debian?
09:59 PM lcnc-relay: <big_kevin420@> or debian was in a different location than rpi...
09:59 PM rdtsc: it's changed a little, think it was in /boot on raspbian
10:04 PM lcnc-relay: <big_kevin420@> you check with linuxcnc yet? did you need to actually use mesaflash>
10:05 PM rdtsc: just using mesaflash to -readhmid
10:05 PM lcnc-relay: <big_kevin420@> if you dont actually need to use mesaflash, i always will say just skip it
10:05 PM lcnc-relay: <big_kevin420@> but i understand if you want to
10:06 PM rdtsc: well I tried running linuxcnc but it crapped out pretty severely
10:06 PM lcnc-relay: <big_kevin420@> and in that case, i still suggest checking to see if linuxcnc works, as a sainity check
10:06 PM lcnc-relay: <big_kevin420@> ok
10:06 PM lcnc-relay: <big_kevin420@> because if you cant get mesaflash AND linuxcnc to work, there probably something wrong
10:07 PM rdtsc: the latest .iso definitely did not have a working spi... hoping that this older version works, since ccatlett84 has got that one to work
10:07 PM lcnc-relay: <big_kevin420@> they wont ever have kernel spi enabled
10:08 PM lcnc-relay: <big_kevin420@> and keep in mind, mesaflash isnt the same as linuxcnc, they dont use the same spi driver
10:09 PM lcnc-relay: <big_kevin420@> so i always suggest checking with both, because if 1 works and the other doesnt, you can waste a lot of time not addressing the correct issue
10:09 PM rdtsc: hmm... could try disabling kernel spi and just running linuxcnc... will have to try that tomorrow, about to try this older sd card real quick
10:10 PM rdtsc: what the heck... tried to write the .iso, look at the sd card, partition types are unknown
10:10 PM lcnc-relay: <big_kevin420@> if you have it enabled in the right file, just leave it, it should work with linuxcnc reguardless of spi setting
10:11 PM rdtsc: it should, but there is a note in the hm2_spix driver docs which say something like "rpi5 users, spi driver switching can be problematic"
10:12 PM rdtsc: https://www.linuxcnc.org/docs/devel/html/man/man9/hm2_spix.9.html#NOTES
10:12 PM lcnc-relay: <big_kevin420@> just use whatever iso you have, i dont think there is a non working spi iso
10:12 PM lcnc-relay: <big_kevin420@> sick i fucking knew it
10:12 PM rdtsc: oh no!
10:12 PM lcnc-relay: <big_kevin420@> ok, one of the things i noticed early on with rpi4
10:13 PM lcnc-relay: <big_kevin420@> was if i ran linuxcnc before i ran mesaflash, mesaflash wouldnt work
10:13 PM lcnc-relay: <big_kevin420@> until i reset the rpi
10:14 PM rdtsc: funky... ok i'll have to pick this back up tomorrow, thanks and 'night
10:15 PM lcnc-relay: <big_kevin420@> hmm i wonder if that note is specifically about trhe spix driver, if he changed a lot from the hm2_rpspi