| - S0rin (QUIT: Ping timeout: 255 seconds) (~S0rin@user/s0rin) | 00:26 | |
| + S0rin (~S0rin@user/s0rin) | 00:30 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 01:11 | |
| - mjw (QUIT: Ping timeout: 255 seconds) (~mjw@gnu.wildebeest.org) | 01:16 | |
| - mtm (QUIT: Ping timeout: 240 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:04 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 02:22 | |
| - qbit (QUIT: Quit: WeeChat 3.7.1) (~qbit@h.suah.dev) | 02:57 | |
| chartreuse | Are the two screw mounts below the CPU for the heatsink soldered in? Just had the right one come off with the screw when I was adjusting the heatsink | 03:02 |
|---|---|---|
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 03:08 | |
| - nsc (QUIT: Ping timeout: 240 seconds) (~nicolas@77-96-142-46.pool.kielnet.net) | 03:08 | |
| + qbit (~qbit@2602:ff16:3:0:1:3a0:0:1) | 03:09 | |
| + nsc (~nicolas@232-97-142-46.pool.kielnet.net) | 03:10 | |
| chartreuse | Guess so, was able to resolder it back down with a bit of effort | 03:24 |
| jackhill | josch: yeah, I ued a computer with 4G or ram for years, but I'll proabably upgrade to the imx.8MPlus module for 8G. It looks like that should have hardware video *encoding* as well! | 03:28 |
| jackhill | like others, more memory for building software. webkitgtk (or maybe just GNOME Web) seens to take up a lot of memory for me. Also, some hi-def large format raster images. | 03:30 |
| jackhill | josch, chartreuse: for getting USB PIDs without $$$$$ https://pid.codes/ | 03:30 |
| chartreuse | Guess either my speakers were loose or from when I had the display not quite all the way tightened down, but now I can actually go over around 65% volume without buzzing XD | 03:40 |
| chartreuse | Still too much treble, but that buzzing issue was the main issue with using the speakers | 03:41 |
| josch | minute: maybe https://pid.codes/ is an option for devices by MNT? | 03:49 |
| josch | minute: my mouse and keyboard only show up as "Atmel Corp. LUFA Mouse Demo Application" and "Atmel Corp. LUFA Keyboard Demo Application" in lsusb, only with the -v flag can i see MNT specific strings -- is that by design? | 03:50 |
| chartreuse | I thought you can change those LUFA strings from saying demo application and such | 03:51 |
| chartreuse | Okay nevermind that's just how the LUFA VID and PIDs are. | 03:52 |
| chartreuse | http://www.fourwalledcubicle.com/files/LUFA/Doc/130901/html/_page__v_i_d_p_i_d.html | 03:52 |
| josch | chartreuse: but that string is sent by the usb device, so it should be changable, no? | 03:57 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:09 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 04:59 | |
| - Boostisbetter (QUIT: Ping timeout: 248 seconds) (4a410829d7@irc.cheogram.com) | 06:08 | |
| + bgs (~bgs@212.85.160.171) | 06:39 | |
| - bgs (QUIT: Remote host closed the connection) (~bgs@212.85.160.171) | 08:07 | |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 08:14 | |
| + andreas-e (~Andreas@2001:861:c4:f2f0::c64) | 09:30 | |
| minute | josch: it is a limitation of lsusb. in dmesg you will see the correct strings | 10:34 |
| minute | josch: lsusb doesn't show the customized strings by default for some reason | 10:35 |
| + mjw (~mjw@45.83.234.184) | 10:43 | |
| josch | minute: i'm still looking for a way to reliably determine that the attached device is indeed some MNT device (and if yes, which). I assume you don't like the idea of https://pid.codes/? Another way would be to set a custom disk label so that the drive would show up in /dev/disk/by-label with a MNT specific name. | 10:44 |
| minute | josch: a drive is only involved for rp2040 trackball, the atmega based trackball and keyboard don't have that | 10:59 |
| josch | ah then i misremember, sorry | 10:59 |
| minute | josch: when the devices are in bootloader mode, i believe there is no way to identify them reliably, in this mode they are just generic mcus | 10:59 |
| josch | hrm... okay, then a rough heuristic via the usb id has to be sufficient | 11:00 |
| minute | i propose that the update is interactive: you ask the user to put the desired device into programming mode | 11:00 |
| minute | and then wait for a flashable device to appear | 11:00 |
| josch | right, that makes sense | 11:00 |
| - qbit (QUIT: Ping timeout: 240 seconds) (~qbit@2602:ff16:3:0:1:3a0:0:1) | 11:18 | |
| + qbit (~qbit@h.suah.dev) | 11:31 | |
| - mjw (QUIT: Ping timeout: 265 seconds) (~mjw@45.83.234.184) | 12:32 | |
| + amospalla (~amospalla@212.231.228.113) | 12:44 | |
| amospalla | Hi, on the pocket croudsupply website, what is the "5G Antenna with Cable & Adhesive Backing" for? Wifi or 4g modem? | 12:46 |
| amospalla | I guess it is for the modem, but better asking. | 12:46 |
| - mark_ (QUIT: Quit: Leaving) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 12:53 | |
| minute | amospalla: it's meant for the modem yep | 12:55 |
| amospalla | thank you | 12:58 |
| minute | josch: or if there are already any flashable devices identified, present the user with choices of which one to flash. my guess is that in 95% of cases people will know what they want to flash. | 13:00 |
| josch | minute: yes, no doubt about that -- probably more like 99% even :) | 13:13 |
| sknebel | is the rp2040-based hardware put into flashing mode the same way as the AVR-based one? (https://mntre.com/reform2/handbook/parts.html#trackball-firmware only describes the latter) | 13:25 |
| - amospalla (QUIT: Quit: Client closed) (~amospalla@212.231.228.113) | 13:35 | |
| - XYZ (QUIT: Remote host closed the connection) (~XYZ@37-48-0-35.nat.epc.tmcz.cz) | 13:40 | |
| + amospalla (~amospalla@212.231.228.113) | 13:44 | |
| amospalla | Jumped in with a Pocket, I can't wait for it to arrive. | 13:45 |
| + XYZ (~XYZ@37-48-0-35.nat.epc.tmcz.cz) | 13:45 | |
| minute | amospalla: awesome, thank you!! | 13:47 |
| minute | sknebel: yep | 13:47 |
| minute | there are also some tricks to force it into programming mode via software, but haven't yet dug into the details of how that works | 13:48 |
| minute | also on recent atmega keyboard fw, you can enter flashing mode from within the menu | 13:48 |
| minute | (i think it's enabled only on the standalone though) | 13:48 |
| sknebel | having to confirm flashing via hardware is not a bad thing IMHO | 13:49 |
| sknebel | although it could of course be something like "hold both buttons to confirm" | 13:49 |
| minute | yeah | 13:49 |
| amospalla | You made a great machine, and the world moves where you put the money. Thank you guys for making such a computer. | 13:50 |
| minute | there's a funny virtual machine escape strategy of reflashing usb devices to keyboards and then entering the magic sequence to exit the VM | 13:50 |
| - XYZ (QUIT: Ping timeout: 276 seconds) (~XYZ@37-48-0-35.nat.epc.tmcz.cz) | 13:50 | |
| minute | amospalla: very much appreciated! | 13:50 |
| sknebel | minute: hah, thats an interesting variant on the flashing risk | 13:51 |
| minute | i think there was such an exploit for the webcam in macbooks | 13:52 |
| minute | that can be reflashed into a malicious keyboard | 13:52 |
| + mjw (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 13:54 | |
| + XYZ (~XYZ@89-24-42-5.nat.epc.tmcz.cz) | 14:02 | |
| - mtm (QUIT: Ping timeout: 255 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:03 | |
| - amospalla (QUIT: Quit: Client closed) (~amospalla@212.231.228.113) | 14:21 | |
| sigrid | looks like g1 on imx8mq does not support scaling in postproc. such a shame :'( | 15:07 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 15:11 | |
| mtm | minute: congrats on exceeding 8-bits in backers on CS! | 15:13 |
| minute | mtm: thanks! we need a 68000 to process these numbers | 15:34 |
| + MajorBiscuit (~MajorBisc@145.94.190.155) | 15:58 | |
| josch | haha :D | 16:33 |
| mtm | looks like exactly 100 more black than purple at the moment. I'm on team purple, ha | 16:37 |
| josch | if i had bought the pocket, it'd definitely be the purple one for me as well :) | 16:37 |
| gsora | purple laptop > any non-purple laptop | 16:39 |
| mtm | absolutely | 16:39 |
| mtm | noticed the ghosting on my OLED (as mentioned in the forum thread); wonder if having an option in the firmware to blank the OLED after no keypress for say, 15 minutes would cure that? | 16:43 |
| minute | clearly this calls for a starfield screensaver | 16:57 |
| gsora | i like how you define your priorities | 17:00 |
| gsora | 11/10 needs a starfield screensaver | 17:00 |
| + bluerise (~bluerise@pc19f8f9b.dip0.t-ipconnect.de) | 17:10 | |
| - andreas-e (QUIT: Ping timeout: 248 seconds) (~Andreas@2001:861:c4:f2f0::c64) | 17:14 | |
| + andreas-e (~Andreas@2001:861:c4:f2f0:b0aa:56dd:ae73:8afa) | 17:14 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 17:40 | |
| - MajorBiscuit (QUIT: Ping timeout: 250 seconds) (~MajorBisc@145.94.190.155) | 17:44 | |
| - XYZ (QUIT: Ping timeout: 255 seconds) (~XYZ@89-24-42-5.nat.epc.tmcz.cz) | 18:02 | |
| + XYZ (~XYZ@37-48-11-232.nat.epc.tmcz.cz) | 18:17 | |
| sevan | https://www.jeffgeerling.com/blog/2023/i-built-special-pcie-card-test-gpus-on-pi | 18:26 |
| minute | sevan: yeah, i believe this is not an issue on layerscape socs | 18:29 |
| minute | (i was using an nvidia card recently w/o major problems) | 18:30 |
| sevan | ack | 18:33 |
| sevan | just impressed by hardware projects. | 18:34 |
| sevan | also, mental picture of more tentacles to extend hardware :) | 18:38 |
| sknebel | https://news.ycombinator.com/item?id=35684859 - article is about imx8, but reforms workaround for the HDMI blob is mentioned a bunch | 18:40 |
| + v4rke_ (~v4rke@88.135.20.160) | 18:44 | |
| - v4rke (QUIT: Ping timeout: 276 seconds) (~v4rke@88.135.20.160) | 18:44 | |
| sevan | I would like to +1 the anti-hollywood sentiment in the comments. | 18:44 |
| josch | and here i am watching my pirated anime (which i'm unable to buy where i live) over a hdmi connection without their hdmi-drm-thing stopping me from doing so... | 18:47 |
| minute | sknebel: weird that this old article surfaces again | 18:50 |
| sevan | reminded my of the drama with Apple and them appeasing the studios. Had to prevent things like itunes from being traceable via dtrace. | 18:50 |
| sknebel | minute: oh, its old? missed that | 18:51 |
| sevan | and the fight with keeping song prices at 99 cents. | 18:51 |
| sknebel | (and doesnt have a publication date on the page now I'm looking...) | 18:51 |
| minute | sknebel: yeah, and it's not about imx8, but imx8mq | 18:52 |
| minute | sknebel: also, imx8mplus does not have a hdmi blob anymore | 18:52 |
| minute | (they use a different, samsung ip) | 18:52 |
| - v4rke_ (QUIT: Ping timeout: 252 seconds) (~v4rke@88.135.20.160) | 18:53 | |
| minute | so on imx8mplus you only have the ddr4 blob. | 18:53 |
| minute | ls1028a + nvidia kepler gpu = blob free system btw | 18:53 |
| minute | (i.e. one that would tick the FSF RYF's checkboxes ^^) | 18:53 |
| josch | personally i don't see much difference between a chip with non-free design and a chip with non-free design that i have to load a binary blob in on startup -- i'm happy as long as i don't have to run a binary blob on kernel/OS level. But we've had this discussion. :) | 18:55 |
| josch | if FSF RYF's checkboxes give more sales to MNT then who am i to judge :D | 18:56 |
| + v4rke_ (~v4rke@88.135.20.160) | 18:58 | |
| andreas-e | josch: Well, one has to draw a line somewhere, and the exact place is a bit arbitrary - as long as we cannot have everything free, from design to user space software, that is. | 19:02 |
| andreas-e | It it helps us run one of the FSF approved distributions on MNT Reform, that would be great! | 19:03 |
| mjw | minute, I would love to see an article about which blobs are still necessary on which MNT (pocket) reform/board combinations and which workaround exist or whether a free replacement can be written or not. | 19:07 |
| minute | we plan to make a table for the pros and cons of the modules | 19:07 |
| mjw | table/article :) | 19:08 |
| mjw | And then there are things like the camera, which has source code but... hmmm :) | 19:10 |
| minute | mjw: hehe yeah. | 19:10 |
| mjw | It is all a bit maddening, we are so close to having a fully free/open system... | 19:11 |
| sevan | joke: for the audiophiles https://audiophilestyle.com/forums/topic/62753-nvme-ssd-designed-for-audiophiles/ (from 2021) | 19:13 |
| + MajorBiscuit (~MajorBisc@2001:1c00:2408:a400:67e:5371:52a7:9b9a) | 19:19 | |
| gsora | as an audio aficionado, seeing this kind of scam products make my blood boil | 19:23 |
| - MajorBiscuit (QUIT: Ping timeout: 260 seconds) (~MajorBisc@2001:1c00:2408:a400:67e:5371:52a7:9b9a) | 19:26 | |
| sevan | gsora: alas this post is missing the golden heatsink that would sit on top | 19:29 |
| josch | andreas-e: that's true. I guess i'm mostly salty about the line that the FSF drew because my favourite OS (Debian) happens to be not free enough for the FSF just because it happens to serve non-free blobs on the same server as the free ones -- even though those non-free blobs are explicitly not part of the Debian distribution. The problem according to the FSF is, that those non-free parts are too easy | 19:31 |
| josch | to access... :/ | 19:31 |
| josch | And for a distribution to be free, it apparently must be really hard to access anything non-free... | 19:31 |
| sknebel | didnt one of the "libre" linux distros patch out the kernel warnings about outdated microcode, because telling users they are running vulnerable microcode could entice them to seek out updates of proprietary software? | 19:32 |
| josch | yeah... somehow your freedoms are more protected by not loading new binary blobs into your intel cpu... | 19:33 |
| sevan | heh | 19:33 |
| sevan | jus an anicdote: when I reflashed my thinkpad x60s with coreboot originally, I opted for no blobs, then realised the CPU wouldn't identify itself as what it was. Loading the microcode made the cpu identify itself correctly. | 19:35 |
| sigrid | meltdown protection, but not the kind you expect | 19:35 |
| + ajr (~ajr@user/ajr) | 19:37 | |
| sevan | look out, it's a hyperthreat! *ducks* | 19:38 |
| - XYZ (QUIT: Remote host closed the connection) (~XYZ@37-48-11-232.nat.epc.tmcz.cz) | 19:52 | |
| + XYZ (~XYZ@37.48.11.232) | 19:55 | |
| - v4rke_ (QUIT: Ping timeout: 255 seconds) (~v4rke@88.135.20.160) | 19:57 | |
| rah | sknebel: no | 19:58 |
| sknebel | no? | 20:01 |
| rah | sknebel: no distros, AFAIK, patched out kernel warnings aobut outdated microcode in order to avoid enticing users to seek out updates of proprietary software | 20:05 |
| rah | sknebel: https://www.fsfla.org/ikiwiki/anuncio/2010-11-Linux-2.6.36-libre-debait | 20:06 |
| rah | "The drivers work as bait, luring users into accepting the deprivation of essential freedoms over the corresponding non-Free Software." | 20:06 |
| rah | nothing to do with updating proprietary software | 20:06 |
| sknebel | https://lists.gnu.org/archive/html/info-gnu/2018-04/msg00002.html | 20:06 |
| rah | sknebel: that post links to the article I just pasted a link to | 20:07 |
| + v4rke_ (~v4rke@88.135.20.160) | 20:07 | |
| rah | sknebel: I see no mention of updating proprietary software in the post either | 20:08 |
| rah | to quote the post: "Since such microcode is non-Free Software, such messages don't belong in GNU Linux-libre." | 20:10 |
| rah | seems pretty reasonable to me | 20:11 |
| sknebel | if providing information about updates to proprietary microcode is considered "software or documentation that induces users to seek and use non-Free programs" (="bait"), then "in order to avoid enticing users to seek out updates of proprietary software" seems like an accurate-enough summary to me | 20:16 |
| sknebel | it's not like the user stops running proprietary microcode just because there is no update installed, they just keep running the vulnerable version | 20:20 |
| sknebel | but this discussion has been repeated often enough, I don't think there is much point in going through it again | 20:20 |
| rah | I think your description was a flippant, malicious dismissal of linux-libre's policies, describing the cribbing of proprietary firmware update messages as "updating proprietary software" as if it's akin to running "apt upgrade" with a bunch of non-free packages and now you're focussing on semantics in order to try and avoid the reality of the essential meanness behind your criticism | 20:24 |
| rah | you wrote "because telling users they are running vulnerable microcode" which shows either a complete misunderstanding of the issue or outright malice in misrepresenting linux-libre's policies | 20:25 |
| Booster[m] | minute: thanks a lot for making the key set for the reform available in the store. I just picked up a set. | 20:46 |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 21:05 | |
| * mark_ -> mjw | 21:05 | |
| + GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 21:05 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 21:06 | |
| - GNUmoon (QUIT: Ping timeout: 255 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 21:07 | |
| minute | Booster[m]: you're welcome! | 21:14 |
| + tretinha (3a571d9f43@2604:bf00:561:2000::1151) | 21:21 | |
| + mtm_ (~Srain@c-71-228-84-213.hsd1.fl.comcast.net) | 21:26 | |
| - frank2 (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~frank2@2a05:f480:2000:1d0d:437d:615e:1650:faf7) | 21:37 | |
| + frank2 (~frank2@2a05:f480:2000:1d0d:563b:b346:faf0:ba68) | 21:37 | |
| - ajr (QUIT: Quit: WeeChat 3.8) (~ajr@user/ajr) | 21:54 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 22:20 | |
| josch | do you guys use gl4es with box64 on your reform? | 22:41 |
| sevan | I have box64 installed as part of getting steam going, how would I check if I'm using gl4es? | 22:50 |
| josch | you would've run your executables with LD_LIBRARY_PATH=~/git/gl4es/lib or similar | 22:52 |
| josch | i'm trying to get Stardew Valley to run on the reform: https://github.com/ptitSeb/box64/issues/742 | 22:53 |
| eery | Why would you use gl4es with native OGL 2.0 support in vivante? | 22:54 |
| josch | i wouldn't -- but i'm not getting it to work so i'm looking for anything to try out | 22:55 |
| sevan | josch: I haven't touched the LD_LIBRARY_PATH but steam is run via a wrapper shell script. | 22:55 |
| eery | I don't think it's GPU related, if it is gl4es should just be worse than native mesa GL, sometimes setting the mesa OVERRIDE env variables to a higher GL/GLSL version helps if it wants a newer GL version | 22:58 |
| eery | Other guess is something in Sid broke box64, it's pretty sensitive to changes from stable | 22:58 |
| eery | I usually run it from a chroot for that reason | 22:59 |
| sevan | josch: nothing relevant in wrapper | 23:00 |
| chartreuse | I can't remember if I got box64 fully working, I know I set up wine for box86, so I think box64 is just left on mine for only 64-bit x86 linux programs | 23:00 |
| sevan | eery: got me wondering if I should patch all the things and see if Thimbleweed Park breaks | 23:01 |
| chartreuse | The whole setup of installing box86 and 64 was a pain IIRC | 23:01 |
| josch | eery: do you have a recipe for that? I think I have to bind-mount /dev to have a graphics card and then also export variables like WAYLAND_DISPLAY or XDG_RUNTIME_DIR=/run/user/1000, right? | 23:01 |
| eery | Well, if you use wayland you'll have to set the environment stuff, if you use Xwayland you just have to set DISPLAY and bind mount /dev and /tmp IIRC, but it's been a while | 23:03 |
| eery | I remember the most annoying bit was programs just silently failing if the GL version was too low :| | 23:04 |
| eery | For me I could get most stuff running, but then it'd just be too slow on the imx8mq | 23:09 |
| minute | josch: you could try SDL_VIDEODRIVER=x11 | 23:28 |
| minute | (or wayland) | 23:28 |
| minute | also, i think there is more detailled logging available | 23:28 |
| minute | also, just anecdotal but i don't need any chroot for box64. | 23:29 |
| josch | neither SDL_VIDEODRIVER=x11 nor SDL_VIDEODRIVER=wayland changed the result -- but good to know that it worked without a chroot! | 23:29 |
| minute | i haven't tried stardew valley though, but i might | 23:30 |
| minute | sometimes i also used strace to get a hint of what the program was doing | 23:30 |
| minute | i.e. to trace file opens and see if something fails | 23:31 |
| josch | the box64 developer ptitSeb is usually super responsive, so they will probably respond to my issue soon and give me the magic rune :) | 23:31 |
| minute | cool! | 23:31 |
| minute | i hope you'll share the eventual solution, i'm curious | 23:32 |
| josch | we ship box64 in the mnt repository -- now i just need to use it myself XD | 23:32 |
| minute | :D | 23:32 |
| josch | minute: sure, i'll drop a line here once it works :) | 23:32 |
| chartreuse | josch: Didn't realize that, I guess I figured when installing it it was grabbing an old version from the debian repo. Though presumably to use either box64 or box86 with wine you still need to build from source with a custom wine as well | 23:37 |
| chartreuse | Getting box86 working was a bit of an annoyance of multilib and such but I have run a windows 32-bit app so it did end up working at least | 23:37 |
| - andreas-e (QUIT: Quit: Leaving) (~Andreas@2001:861:c4:f2f0:b0aa:56dd:ae73:8afa) | 23:58 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!