jaeger changed the topic of #crux-arm to: CRUX-ARM 3.6 Released! - http://crux-arm.nu/Documentation/ReleaseNotes3-6 | Logs: https://libera.irclog.whitequark.org/crux-arm/
genpaku has quit [Remote host closed the connection]
genpaku has joined #crux-arm
<mnkydeth> ok, back after work. I made more progress. I have the cmdline.txt and config.txt file made and populated. But ran into another issue. And just for reference this is what I have in my cmdline.txt
<mnkydeth> console=tty1 root=/dev/mmcblk0p3 rootfstype=ext4 rootwait ro quiet
<mnkydeth> But, upon putting the sdcard into the IO Board, and booting the CM4 it still thinks root is at /dev/mmcblk0p2 and not /dev/mmcblk0p3
<mnkydeth> It even notifies me that /dev/mmcblk0p2 is a swap partition. I can go back and do two partitions. Maybe setup a swapfile instead of an actual partition for swap.
<mnkydeth> For giggles I tried with two partitions but instead of using ext4 I used f2fs. My cmdline.txt is console=tty1 root=/dev/mmcblk0p2 rootfstype=f2fs rootwait ro quiet
<mnkydeth> So, it seems it may not be reading this file?
<mnkydeth> Now to do the two partitions and use ext4
<mnkydeth> Doing that, two partitions not three, and using fat32 and ext4 works but not f2fs.
<mnkydeth> Ahh, and the usb port on the IO Board doesn't work. I'll need to look up how to enable that.
<jaeger> Did you also update /etc/fstab to point to the right root?
<mnkydeth> I did not. I'll check that in a few to see if I can use f2fs.
<mnkydeth> But it still didn't allow the usb ports to work on the IO Board.
<jaeger> Not sure there, I've never used a CM
<mnkydeth> As I have a keyuboard in the USB ports. And I know this is not a Raspberry pi4 but it is a CM4 on an official IO board.
<mnkydeth> Gonna do a copy pasta
<mnkydeth> Just a comment to keep someone else from falling into the same pit I did. I added this line to config.txt by connecting it via J11 and letting boot as a USB storage device. Once I could see it in Windows I edited config.txt. But no matter what the USB mouse and keyboard would never work. Finally I realized I still had power applied to J11 via the microusb cable. I recall from the schematic this flips a
<mnkydeth> switch that re-routes the USB signals from the HUB and instead to J11. Unpluged from J11 and all if FINE! Only 20 minutes lost on this. :-) Not unusual for unfamiliar hardware. Lesson learned. Hope this helps someone else not waste 20 min or more.
<mnkydeth> So, this was my fix for the usb ports on the IO Board.
<mnkydeth> usb keyboard works
<jaeger> nice
<mnkydeth> I know a lot of this stuff is noise. But I kind of at least want it in the irc log in case anyone else uses this and runs into these issues.
<mnkydeth> Will help with documentation down the road as well I think.
<mnkydeth> Now, just for my reference. I know I used a RPI4 base. Is there a generic aarch to use? So I could try on my Pi zero 2 w?
<mnkydeth> Or an optimized base for that device?
<jaeger> Not sure if I have one, will check. I have the rpi4-optimized one still but don't know if I saved a generic one
<jaeger> pitillo probably has one
<mnkydeth> Well, I'm not sure how I can help per say as I know you guys are still ironing out the base for a release right? I figured testing and giving what I go through might help in some way.
<mnkydeth> But, I'm very interested in using Crux on my Pi's. And I plan on setting up a bit of a distcc farm here for it.
<jaeger> More testing never hurta
<jaeger> hurts
<jaeger> Yeah, looks like I did not save my generic rootfs, sorry
<mnkydeth> It's all good. I appreciate you looking.
<mnkydeth> I know this isn't release material yet either, otherwise it would be up on the crux-arm page.
<jaeger> yeah
<jaeger> I think there are still some bugs to iron out with the crux-arm-release repo
<jaeger> but I've started a generic build to see how it goes
<jaeger> Another option if you don't mind taking a lot of time is to install a generic 3.6 release and then rebuild the toolchain and go from there
<mnkydeth> I am very green when it comes to building into a new release. I've done LFS a few times back in the day and never had much success. I honestly work on things very slowly anyways since my wife and kids always want me to do stuff when I'm home and I get 5min here and there to poke on my computers. But, if there are instructions on how to do so I could attempt to get a pi Zero w and pi zero 2 w image made.
<mnkydeth> So, it may not be for me unless there are specific instructions.
<jaeger> It's usually not too tough, just very time-consuming
<jaeger> Up to you, happy to help if you want to go that route, or you could wait for a 3.7 generic rootfs
<mnkydeth> Well, if I do this with a pi zero 2 w what is the typical sdcard size I will need? I have up to 512MB here. No ssd's for the pi zero's so.... To bad they didn't have more ram for ram compiling.
<mnkydeth> Hmm, there's nothing in boot on the 3.6 release that I see. I'll need more then just the two files right?
<mnkydeth> Might be best to wait for a generic 3.7 release. lol
<mnkydeth> Just an fyi .... I put the sdcard with the 3.7 pi4 optimized base. Into a pi zero w 2 and it booted all the way to the command line.
<jaeger> I usually buy 16GB or 32GB uSD cards since they're pretty cheap
<jaeger> nice
<jaeger> ok, I uploaded a generic aarch64 rootfs to https://crux.ninja/tmp/
<jaeger> Not tested
<mnkydeth> Thanks, I'll grab it and try it out when I can.
<beerman> jaeger: are you able to post your output of `chronyc sources`? I am especially interested in your NMEA entry
<beerman> also i just noticed that it also outputs a pps device, so it could be even more accurate 🤓
<beerman> at least in theory, following another blogpost that you posted
<pitillo> sorry, I'm following all your talks from the shadow (some crazy weeks here at work and very tired latelly)
<pitillo> mnkydeth: here there isn't noise, everything is welcome... respecting xD
<pitillo> there isn't too much voices to disturb with any topic
<pitillo> btw, we host all work done at sourceforge... genereic releases for 32b/64b and optimizations
<jaeger> I couldn't remember if there was one there for 3.7
<pitillo> a bunch of them...
<pitillo> I miss yours for rpi4 on SF
<pitillo> let's see if I can add it tomorrow morning from home
<beerman> oh perfect, i forgot you posted this
<jaeger> No worries, I just went looking for it as the pi is off right now
<beerman> whats your antennas situation? it tracks me very well, and its very close to the roof placed on a metal plate. I get 15 connections but chrony thinks my clock is a falseticker
<beerman> x indicates a clock which chronyd thinks is a falseticker (i.e. its time is inconsistent with a majority of other sources).
<beerman> #x NMEA 0 3 17 3 +38ms[+1955us] +/- 1000us
<beerman> i want to think my antenna is doing fine, if i read gpsmon correctly
<jaeger> I'm using the tiny antenna that came with the module as I didn't have my own... it's close to an outside wall in my office, nothing special
<jaeger> Does gpsmon have multiple sats connected?
<beerman> i forgot i asked that already as well. Fine 😄
<beerman> 15
<beerman> thats the upper left part of the view, right?
<jaeger> It'll list all of the ones that it has seen in the left if I remember right, on the right it will say how many are actively locked
<jaeger> If you can post a screenshot I can point out where
<jaeger> hrmm... that looks a little different from what I recall.. maybe a version difference
<jaeger> I'm guessing the ones with the U flag showing Y are the active ones but I could be wrong. I'll move my pi over and start it up to see
<beerman> looking at the internet and at the blog post: yep
<jaeger> This is mine before it gets any locks
<jaeger> Sats: 00
<jaeger> That'll update to whatever number is active in a bit
<jaeger> I haven't seen the x from chronyc sources yet, just works fine after it gets some signals
<jaeger> Now it says Sats: 04 and chrony switched to it automatically
<jaeger> #* NMEA 0 3 377 6 +928us[+1326us] +/- 1000us
<jaeger> That number matches the number of sats in the left list with the 'Y'
<beerman> alright, so the antenna works
<beerman> i ditched the other sources for a while
<beerman> seems to vary between 6 to 8
<beerman> i reenabled a few pools and its immediately #x again.. weird
<jaeger> wrong offset, maybe?
<beerman> mh, maybe
<beerman> that offset is varying awfully much it seems...
<jaeger> I didn't watch my variation closely but I gathered a lot of tracking samples and took the average
<beerman> thats what i tried to do, yeah
<beerman> ok, yeah, offset was wrong, dunno how i did that wrong with libreoffice, but R to the rescue
<beerman> or not? lol, its x again
<beerman> i will probably try to move it a bit, later
<beerman> the pps does not appear working either
<jaeger> I haven't tried to use pps yet
<pitillo> uploaded rpi4-rootgs to sourceforge too...
<jaeger> nice
<beerman> very nice indeed