2026-02-03.log

- AnimaInvicta (QUIT: Ping timeout: 240 seconds) (~AnimaInvi@88-120-179-216.subs.proxad.net)00:06
+ AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net)00:08
- chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)01:06
- AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net)01:12
joschthank you, i suggested that to a user on the forum, lets see if it works01:44
- burley (PART: #mnt-reform) (fc32d5b940@user/burley)01:48
- mjw (QUIT: Ping timeout: 256 seconds) (~mjw@gnu.wildebeest.org)01:50
- pomel0 (QUIT: Ping timeout: 264 seconds) (~pomel0@user/pomel0)02:22
+ pomel0 (~pomel0@user/pomel0)02:22
bremnerthe good news is that I didn't short anything removing the SSD from my pocket reform. The bad news is that the hibernate / idle tests are still failing02:27
- paperManu (QUIT: Ping timeout: 260 seconds) (~paperManu@146.71.9.156)02:44
- kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de)03:06
+ kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de)03:08
- kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de)03:08
+ kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de)03:08
- pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0)03:09
+ pomel0 (~pomel0@user/pomel0)03:10
- pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0)03:27
+ pomel0 (~pomel0@user/pomel0)03:27
wickedshellI haven't been able to keep up here recently, but my pocket reform has developed a behaviour of losing keyboard and trackball input. Resetting the keyboard doesn't seem to work, but resetting the USB controller does sometimes. I recently had gnome firmware window open when it happened again and it showed the keyboard device completly disappearing. Both system and keyboard appears to be fully up to date. (Not as sure on the ke03:32
wickedshellyboard, but it was like 4 days ago).03:32
wickedshellAbout to take it apart and start looking, but curious if anyone else already hit all of this.03:33
- paperManu_ (QUIT: Ping timeout: 240 seconds) (~paperManu@146.71.9.156)04:09
- voltaire28_ (QUIT: Ping timeout: 246 seconds) (~jlafon@28.162.2.93.rev.sfr.net)04:27
+ voltaire28_ (~jlafon@28.162.2.93.rev.sfr.net)04:29
- voltaire28_ (QUIT: Remote host closed the connection) (~jlafon@28.162.2.93.rev.sfr.net)04:29
wickedshellWelp, after finding nothing wrong, reassembling it seems to be behaving better, which is absolutely the worst outcome in my opinion04:37
- boonet (QUIT: Ping timeout: 250 seconds) (~boonet@user/boonet)06:05
- pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0)06:24
+ pomel0 (~pomel0@user/pomel0)06:36
joschbremner: i put a few more pcie suspend related commits onto 6.19~rc6 -- did you try it out?07:05
joschwickedshell: there was another user recently who has reported that their usb-cable managed to wiggle free and that re-seating fixed their problems07:05
bremnerjosch: not yet, thanks for the headsup07:10
bremneractually I was not even using the debian kernel...07:10
joschbremner: i packaged 6.19~rc6 from experimental in this MR: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/16607:31
joschmore recent linux is currently waiting in NEW07:31
bremnerjosch: so I need to build it?07:45
joschbremner: no, just add this apt repo: deb [trusted=yes arch=arm64] https://source.mnt.re/reform/reform-debian-packages/-/jobs/18097/artifacts/raw/repo reform main07:48
joschbremner: you find that last line in the "reprepro" job log at the bottom07:51
joschgordon1: after they have forgotten about me, sprintek now shipped the parts even though they didn't receive my payment yet. :D09:40
joschrelated: signing up with wise.com is a nightmare... lets give them 100 selfies of mine, government id photos, tax id, phone number, employer, addresses...09:41
+ chomwitt (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)10:11
- timide (QUIT: Read error: Connection reset by peer) (~timide@user/timide)11:50
+ timide (~timide@user/timide)11:51
+ mjw (~mjw@gnu.wildebeest.org)11:59
- chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)12:02
- jogu (QUIT: Remote host closed the connection) (~jogu@user/jogu)12:21
+ jogu (~jogu@user/jogu)12:46
- jogu (QUIT: Remote host closed the connection) (~jogu@user/jogu)12:48
+ jogu (~jogu@user/jogu)12:49
+ paperManu (~paperManu@146.71.9.156)12:51
minutejosch: employer? :013:19
joschnope, buying a trackstick for the mnt keyboard trackstick mod from sprintek.com :)13:20
+ boonet (~boonet@user/boonet)13:20
minutejosch: no i mean, they ask about your employer @ wise?13:20
joschyes, they do13:20
joschi mean you can enter anything but still...13:21
gordon1josch: they trying to be nice now, maybe they even put a candy in your package13:23
minutejosch: ah maybe because it's not a business account, the KYC stuff is different13:23
joschhaha lets see :)13:23
minutejosch: btw just a heads up, i briefed ailurux about the next todos for barebox, namely to clean up the code for release/mainlining and introduce all the correct DTSes for rk3588 devices13:29
minutejosch: not sure how long it'll take, maybe 1-2 weeks13:29
minutejosch: (he works part time)13:29
joschvery nice!13:30
minutejosch: it has been stable for me so far now on pocket reform, but the dts is a mutant of reform2 and pocket, because it grew out of the reform2 hdmi test13:30
minute(the one that barebox uses i mean, not the one that's loaded for linux)13:30
minuteand it would be interesting to turn on simple framebuffer support + test some efi images :313:31
minutei mean like fedora or openbsd etc13:31
joschthe tool "reform-flash-uboot" was a misnomer from the start because it doesn't just flash u-boot but also the bootloaders before u-boot. Maybe a rename with a compatibility symlink should be created and the tool should actually be called: reform-flash-bootloader?13:31
joschi already booted fedora13:31
minutejosch: oh that sounds good yeah.13:31
minutejosch: ah, but with fedora's kernel?!13:31
joschyes, so display does not work13:32
joschi need somebody familiar with fedora who knows how the fedora kernel packaging works13:32
minutejosch: cool, but it might work if barebox is configured to pass the simpleframebuffer13:32
joschmaybe, it didn't work on the pocket though13:32
minutejosch: yeah it's not configured yet, that's what i mean :D13:32
joschah okay13:32
joschsorry :D13:33
joschyes, lets try that out :)13:33
minutethere is CONFIG_DRIVER_VIDEO_SIMPLEFB13:33
minute> Add support for setting up the kernel's simple framebuffer driver based on the active barebox framebuffer.13:34
- buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org)13:34
joschanother thing that is needed is that barebox orders the kernels correctly13:34
minuteah, it sorts wrongly?13:34
joschcurrently, when both 6.18 and 6.19 bootspec files are there, barebox will load 6.1813:34
minutesounds like the sort order should be flipped13:35
joschyes, but a3f read about that and is fixing that13:35
+ buckket (~buckket@vps.buckket.org)13:35
minutecool13:35
joschjust noting it as another thing that should be there before making it official13:35
bremnerjosch: behaviour of experimental kernel seems similar to rc7 (I guess that's where you got the patches from?) platform test claims to be waiting for 5s but never comes back13:36
minutedrivers/video/simplefb-fixup.c is quite readable, interesting how that works13:37
minuteit adds a dts node on the fly13:37
joschbremner: i got my patches from collabora rockchip-devel13:37
minutewith compatible "simple-framebuffer" and also marks the memory as reserved13:37
minuteand passes width, height, stride and format, and of course the address in memory to the kernel in this dts node13:38
minuteso seems straight forward13:38
bremnerjosch: ok. that is indeed where I built my kernel from13:40
bremnerI'll test a bit more later13:40
a3fminute, I haven't tried it on an RK3588, but on i.MX93, you'll need some extra properties in the framebuffer node, so simpledrm knows which power-domain and clocks to keep active. you can also reference a panel this way, e.g. for brightness control13:48
a3fbarebox will reuse an existing /framebuffer node if it exists, so you'll need to either populate that in the kernel DT or register a fixup in board code13:49
minutea3f: ah, very good to know, thank you!14:04
minutebtw i'm working on pocket motherboard v1.5 at the moment, and will do some small power related changes cc ch14:04
minutech: i'm replacing D11 with an R0, and will put in 2x AP22615A for the USB-C ports instead of 1x AP22815A for only the charging port. AP22615A have a resistor set current limit.14:05
chah that sounds cool14:06
chplus an id rom or something so the firmware can know?14:06
minutech: yes, that's on my list as well, i'm thinking what's the best way to do that to also make that easy during production14:07
minutemaybe in rp2040 fw something like, "if id eeprom is empty, populate with this default"14:08
- paperManu (QUIT: Ping timeout: 255 seconds) (~paperManu@146.71.9.156)14:09
minutenow the question, what should be the hard limit on our USB-C port power outputs14:09
minuteor should i servo that iset resistor (a bit more complex though)14:09
minutech: ah wait, main 5v doesn't need to be switched to the charging port VBUS output at all... because the charger module OTG mode can do that independently, right?14:15
chah yes14:18
chso that can all go14:19
chminute: some complexity with the idrom: it needs to be possible to be read soon after startup so charging from empty batt can work14:20
minutech: ok14:24
- pomel0 (QUIT: Ping timeout: 260 seconds) (~pomel0@user/pomel0)14:25
+ pomel0 (~pomel0@user/pomel0)14:29
- kensanata (QUIT: Quit: OK) (~alex@user/kensanata)14:40
+ kensanata (~alex@user/kensanata)14:41
+ paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca)14:44
- mjw (QUIT: Ping timeout: 245 seconds) (~mjw@gnu.wildebeest.org)14:47
* Guest8268 -> mjw14:47
+ chomwitt (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)14:55
- chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)15:08
+ spew (~spew@user/spew)15:28
- paperManu (QUIT: Ping timeout: 244 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca)15:31
+ chomwitt (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)15:34
+ paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca)15:39
- pomel0 (QUIT: Ping timeout: 260 seconds) (~pomel0@user/pomel0)15:46
+ pomel0 (~pomel0@user/pomel0)15:46
joschminute: Do you consider pcie_aspm=off just a temporary workaround or something which should be added to the defaults?15:50
- pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0)15:59
+ pomel0 (~pomel0@user/pomel0)16:00
minutejosch: hard to say. temporary until all pcie power issues are fixed in the kernel... don't know how long that'll take16:02
minutejosch: ideally we should always be able to tweak kernel options via updates without needing to update the bootloader16:02
joschminute: we can do that for people who use u-boot-menu to boot instead of boot.scr16:08
joschwe can create /usr/share/u-boot-menu/conf.d/reform_pcie_fix.conf and put into it:16:09
joschU_BOOT_PARAMETERS="$U_BOOT_PARAMETERS pcie_aspm=off"16:09
joschand once this is fixed in the kernel, the next reform-tools version drops that file16:09
minutech: i came up with a power management strategy. i assume our peak input budget from the batteries is 4A @ 7V. there should be around 1.6A always reserved for the SoC (~11W TDP afaik). so my plan is to limit the 3.3V peripheral rail to 1.5A of the 7V input. and limit the USB 5V output total to 0.7A. that leaves a bit of headroom for 3v3 standby. 16:10
- _justin_kelly71 (QUIT: Quit: The Lounge - https://thelounge.chat) (~justinkel@user/justin-kelly/x-6011154)16:11
minutei'm thinking to put USB vbus output for the second usb-c port downstream of the main soc 5v as before, but as mentioned behind a switch with limiter so we can set that to 0.5A or so. if i have extra time i could put a fet there that would change the limit resistor. but not sure yet16:11
minutejosch: ok, sounds good!16:12
minutejosch: i think everybody should move to u-boot-menu, no?16:12
joschminute: there is one downsied: this will add this option for *everybody* not just rk358816:13
minutejosch: what's the downside?16:13
joschi have no idea whether pcie_aspm=off has any negative effects on other platforms16:13
minutei'm pretty sure pcie aspm leads to all sorts of issues as it's often broken16:13
+ _justin_kelly71 (~justinkel@user/justin-kelly/x-6011154)16:13
minutejosch: it will always lead to more stability, it could just lead to more power draw in case aspm worked for someone miraculously before16:14
minutejosch: but then, apparently we had clk_ignore_unused all this time which probably also caused some power waste16:14
joschokay, but i'll open an issue in bugs/bugs/-/issues about this and reference that as a comment in /usr/share/u-boot-menu/conf.d/reform_pcie_fix.conf okay?16:14
minutejosch: yep16:14
joschthen we'll not loose track of this in the future when we wonder why we did this in teh first place :D16:15
minutejosch: good call!16:21
- chomwitt (QUIT: Ping timeout: 244 seconds) (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)16:55
elb`I think that pcie bus reconfiguring message is not spurious17:00
elb`I just had it get to the booting forky bit, then it printed that message, and boot hung17:00
elb`first time I've seen that17:00
elb`and then it just did it again17:01
elb`hmmm an updated kernel is available, hopefully that helps17:10
+ bkeys (~Thunderbi@66.110.201.50)17:12
- pomel0 (QUIT: Ping timeout: 265 seconds) (~pomel0@user/pomel0)17:39
+ pomel0 (~pomel0@user/pomel0)17:40
+ mark_ (~mjw@gnu.wildebeest.org)17:42
+ bkeys1 (~Thunderbi@129.sub-174-229-17.myvzw.com)17:49
wickedshelljosch: yeah I was hoping to find a obviously unseated cable. Everything looked fine unfortunately. I pushed all the connectors to make sure they were fully seated, but it didn't seem to move at all. I was checking for wear against the hinge as well and didn't find anything there17:50
wickedshellBut no problems all evening after opening it up and putting it back together, so I guess I shouldn't complain. I hate phantom "fixes" though17:51
- bkeys (QUIT: Ping timeout: 265 seconds) (~Thunderbi@66.110.201.50)17:53
* bkeys1 -> bkeys17:53
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@129.sub-174-229-17.myvzw.com)17:58
+ bkeys (~Thunderbi@129.sub-174-229-17.myvzw.com)17:58
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@129.sub-174-229-17.myvzw.com)18:03
+ bkeys (~Thunderbi@129.sub-174-229-17.myvzw.com)18:05
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@129.sub-174-229-17.myvzw.com)18:10
+ bkeys (~Thunderbi@129.sub-174-229-17.myvzw.com)18:11
- spew (QUIT: Quit: nyaa~) (~spew@user/spew)18:47
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@129.sub-174-229-17.myvzw.com)18:55
- casparvitch (QUIT: Read error: Connection reset by peer) (~casparvit@user/casparvitch)19:06
+ casparvitch (~casparvit@user/casparvitch)19:06
- jogu (QUIT: Ping timeout: 260 seconds) (~jogu@user/jogu)19:07
- schalken (QUIT: Ping timeout: 240 seconds) (~schalken@117-118-178-69.gci.net)19:08
+ schalken (~schalken@117-118-178-69.gci.net)19:10
minutei'm making charger bypass automatic and eliminating the internal usb-c port and the jumpers, instead there's a cascade of usb2 signal switches19:21
minuteto the usb2 pins of second outside usb-c port19:22
minuteso you can select what you want there, usb hub, usb2 host (direct) from soc, or rp204019:22
minuteso then rp2040 flashing and diagnostic can be done without opening the device19:23
minuteelb`: you already have pcie_aspm=off ?19:23
joschwickedshell: yeah, very anticlimactic and unsatisfactory that :(19:28
joschwow, big pocket reform motherboard changes o019:29
minuteyeah19:29
- pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0)19:31
+ pomel0 (~pomel0@user/pomel0)19:31
minuteput AT24CS02-STUM as a eeprom with unique serial number19:47
minutecould also use a bigger version for saving preferences or sth :D19:48
- chrcav (QUIT: Quit: Lost terminal) (~chrcav@user/chrcav)19:51
* mjw -> Guest2920:42
- Guest29 (QUIT: Killed (uranium.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d)20:42
* mark_ -> mjw20:42
+ Guest29 (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d)20:43
+ xktr (~xktr@user/xktr)20:45
+ chrcav (~chrcav@user/chrcav)20:47
+ siviq (~siviq@user/siviq)20:53
- siviq (QUIT: Quit: Client closed) (~siviq@user/siviq)21:02
+ bkeys (~Thunderbi@172.56.67.154)21:08
elb`minute: I do not, is that likely to affect nvme as well as wi-fi?21:45
+ voltaire28_ (~jlafon@28.162.2.93.rev.sfr.net)21:45
elb`for some reason I thought it was wi-fi only21:45
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@172.56.67.154)21:54
+ bkeys (~Thunderbi@172.56.67.154)21:54
+ chomwitt (~chomwitt@2a02:85f:9ac6:8200:1ac0:4dff:fedb:a3f1)22:02
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@172.56.67.154)22:24
+ bkeys1 (~Thunderbi@172.56.67.154)22:24
* bkeys1 -> bkeys22:27
minuteelb`: pcie in general22:47
minuteelb`: give it a shot22:47
minuterare pocket reform with bonded glass/display https://mastodon.social/@mntmn/11600886628872965422:47
elb`yeah I'll try22:54
* elb` -> elb22:54
elbshiny22:55
joschjust for the visuals or for any practical purpose?22:56
- paperManu (QUIT: Ping timeout: 256 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca)23:03
- bkeys (QUIT: Ping timeout: 255 seconds) (~Thunderbi@172.56.67.154)23:05
+ paperManu (~paperManu@146.71.9.156)23:18
+ paperManu_ (~paperManu@146.71.9.156)23:19
- paperManu_ (QUIT: Ping timeout: 244 seconds) (~paperManu@146.71.9.156)23:23
- pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0)23:34
+ pomel0 (~pomel0@user/pomel0)23:35
+ paperManu_ (~paperManu@146.71.9.156)23:36
minutejosch: mostly aesthetics23:57

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!