Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2024-10-11

00:00:21 Quit othello7 (Ping timeout: 248 seconds)
00:02:48 Join baltazar [0] (~baltazar@user/baltazar)
01:00
01:45:39***Saving seen data "./dancer.seen"
02:00
02:52:35 Join WebGuest60 [0] (~WebGuest6@2c0f:f4c0:2202:1fd8:f16f:a097:c33f:785f)
02:52:40 Quit WebGuest60 (Client Quit)
03:00
03:45:40***Saving seen data "./dancer.seen"
04:00
04:22:33 Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net)
05:00
05:04:29 Quit jacobk (Ping timeout: 260 seconds)
05:04:45 Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net)
05:07:30 Quit bpye (Ping timeout: 248 seconds)
05:45:41***Saving seen data "./dancer.seen"
07:00
07:45:43***No seen item changed, no save performed.
08:00
08:14:09 Part spork
08:17:55 Join spork [0] (topic@77-164-207-8.fixed.kpn.net)
09:00
09:26:15speachydconrad: Interestingly, NET support was actually added in the hw1.5 kernels, wasn't present in the hw1 kernels.
09:27:24speachythere's differences between hw1 and hw1.5 in the AXP supply configuration. Not sure if that's something we need to care about or if the hw just comes up in the proper configuration.
09:28:09speachy...I added a hw1 dmesg to the wiki page.
09:35:27speachyaos, did you have a chance to try the hosted HW vol patch for hw2+ ?
09:35:47speachys/aos/also/
09:35:49speachy...
09:45:44***Saving seen data "./dancer.seen"
10:00
10:03:50 Quit danwellby (Read error: Connection reset by peer)
10:06:25 Join danwellby [0] (~danwellby@cynthiajndevey.plus.com)
10:19:53 Join dconrad [0] (~dconrad@152.117.104.217)
10:20:39dconradspeachy: I haven't yet, no. I realized that I'll need to get the hosted compiler running first, I don't think I got my docker image to build it correctly
10:20:51speachyI can toss you a build if you like.
10:21:20dconradyeah, that would be good - I was planning on adding a debug menu line as well just to see if it is indeed doing what we think its doing
10:22:03speachydconrad: https://www.shaftnet.org/~pizza/erosq-hw2test.zip
10:22:36speachy(It's what's running on the hw1 in front of me)
10:23:21dconradah, perfect. I'll give it a spin
10:26:42dconradI also did some formatting on the wiki page, it had gotten a little hard to read I thought
10:26:58speachyyeah, it's grown a lot..
10:29:33dconradhaha I think its pretty clear it's _not_ working, I just get full-scale output regardless of the volume
10:29:52dconradAlso seems to be missing the theme backdrop, which is odd
10:29:52speachyheh, clearly someting wrong then. :D
10:30:20speachythat might be due to LO detection..
10:30:23dconradthe output seems to switch between hp and lo correctly though
10:30:45dconradoh, let me try setting the maximum volume setting
10:31:37dconradnope, still blowing out my speakers
10:38:02speachythis should be straightforward to fix.
10:38:19speachyJust added debug menu reporting for hwver too
10:38:26speachy1/23/4
10:53:43speachydconrad: updated the zip, try again? it'll probably break the same way, but I'm curious what the debug menu says.
10:54:50dconradok, let's see
10:57:16dconradthe hwver field is changing between 22 and 23 at about 1hz
10:57:56dconradhp and lo readouts accurately reflect the state it seems like
10:58:17speachyaha, found the reason for the 22>23.
10:58:46speachyaudio still behaves the same way?
10:58:54dconradyep, audio is still at max
10:58:56speachyk.
11:00
11:06:07dconradI didn't realize so much information was available in the dmesg readout, if the LCD hasn't changed from hw3 I might not need to decompile the OF... it seems like it's 3 additional gpio pins and getting the axp2101 working
11:06:24dconradhopefully I can work on that this weekend
11:07:20speachythere's no useful LCD info in any of the dmesgs
11:07:33speachybut it may be possible to probe around in the sysfs stuff to find more info
11:07:58speachybut yeah, we can query REG03 of the AXP to find out what it is, and adjust what we do from there.
11:08:10dconradI think my plan is "ignore the lcd right now and it'll be obvious in a bit" haha
11:08:50dconradthe way the code's setup assumes the AXP device type is hardcoded, but I think that can be changed without major overhaul
11:09:15 Quit jj5 (Ping timeout: 260 seconds)
11:09:22speachythe linux AXP PMU driver might help as a reference
11:17:32dconradalright, I'm going in on a surfans, whether I get a hw3 or hw4 it's one I don't have yet
11:17:38dconradlet's hope 4
11:34:04 Join othello7 [0] (~Thunderbi@pool-100-36-176-164.washdc.fios.verizon.net)
11:36:12speachyfingers crossed.
11:45:45***Saving seen data "./dancer.seen"
11:46:48speachyI wonder if one has to muck with the output selection switch differently on HW2+ for volume to work.
11:48:06dconradI doubt it, you can see my ascii art in wiki - on hw1/1.5, the LO is direct off the DAC and HP goes through both HP amp and selection switch
11:48:40dconradon hw2/3, the LO is changed to be on the other output of the selection switch
11:48:56dconradand I was getting output on both HP and LO outputs at max volume
12:00
12:01:55speachythat "output selection switch" is just a construct of their linux driver; it maps a numeric value into whatever gpio incantations needed to ake the desired output work.
12:02:38speachyupdated the zip again
12:04:27dconradok, I'll give it a shot
12:05:16tertuoh i imagine this question has come back a few times before, but what's the best way to compile gcc 4.9.4 if you do end up needing to compile it
12:05:52 Join bpye [0] (~bpye@user/bpye)
12:06:08speachytools/rockboxdev.sh
12:06:27tertuMy current thought is run the tools build script until it fails, delete the build dir, patch gmp's configure script (which... I'm a bit annoyed at what they were doing! they could have just done return 0 even back then), and then re-zip the gcc 4.9.4 archive
12:06:59tertugmp's configure script includes a number of test programs of the form int main() { exit(0); } or something like that
12:07:04tertuand modern compilers don't like it
12:07:15speachywhat's "modern" in this context?
12:07:26tertuwhatever the version of gcc i have installed is? 14 i think?
12:09:31speachyI'm 98% sure I tested GCC14, but I kicked off another build; let's see.
12:09:35tertuIt won't because iirc the config script uses -Werror at this point for some reason and the compiler warns on an implicit declaration of exit()
12:10:01tertuthis is on aarch64 linux if that matters
12:11:00tertu(i'm on macos right now so I can't test it, unfortunately, but i can reboot into linux later)
12:11:24 Part chamlis
12:11:32dconradspeachy: volume control works now, but for 2 issues: it doesn't seem to send the volume until you change it, and we get some nasty clipping even at low volumes
12:12:21speachyif you have a patch to gmp or whatever, rockboxdev.sh can apply it automatically
12:12:42dconradand I think we had the volume limit being applied to LO, this seems to have broken that
12:13:04tertuyeah the config script fix is pretty simple, you just find every "exit(0)" and replace it with "return 0", I'll see if it needs more patches past that
12:13:24tertuit's just the basic "does the compiler work at all" variety of test
12:14:06speachytertu: would re-running autoconf do the trick instead?
12:14:14tertuhmmmmm maybe
12:14:38tertui'll give that a shot later
12:15:26speachy... this went boom on me but no visible error/failure (in the gcc configure script)
12:15:51speachyah here we go.. bcopy, in gmp.
12:17:20TrzyzetI have F20 with USB-C, maybe I can help with something?
12:20:43dconradI've got to take off in a half hour or so here, I'll be back tonight or tomorrow
12:30:15 Quit dconrad ()
12:36:59speachynope, the problem is in the macros. I'm inclined to just bump it up to GMP 6.1 as that was the latest at the time of GCC 4.9.4's release
12:42:19tertuisn't gmp bundled in the 4.9.4 source tree though?
12:43:04speachynope.
12:43:39speachywe _could_ punt to using the host's gmp/mpfr/isl/mpc like we do for the hosted toolchains.
12:44:29speachyok, bumping gmp to 6.1.2 (4.3.2) and mpfr to 3.1.6 (3.1.0) seems to have done the trick.
12:46:23speachyI wonder why I settled on 4.3.2 given how old it was even at the time...
12:46:56speachynow for the other two, make sure those still build cleanly.
12:52:46 Join chamlis [0] (~chamlis@user/chamlis)
13:00
13:01:57speachy4.3.2 was the minimum required version for GCC 4.9.4, apparently.
13:02:06rb-bluebotBuild Server message: New build round started. Revision 49b6853c19, 345 builds, 10 clients.
13:02:06rb-bluebotrockboxdev.sh: bump required version of GMP and MPFR by Solomon Peachy
13:02:18speachytertu: grab the update, and try again.
13:03:03speachyand thanks for the report; I could have sworn I'd tested this on a GCC14 prerelease.
13:07:36speachydconrad: methinks I should get my hands on a hw2/3 &| hw4 unit too.
13:07:57 Part chamlis
13:12:18bluebrothersdl2-config −−static-libs appears to be somewhat broken on Debian stable
13:12:45bluebrotherit expects CC to be set which usually isn't the case, then returns an invalid path to libSDL2.a
13:13:21bluebrotherwhich in turn raises the question: why are we trying to link statically for sim builds anyway?
13:14:04bluebrother(it's only an issue when trying to get static linker flags)
13:14:36 Join chamlis [0] (~chamlis@user/chamlis)
13:15:26 Part chamlis
13:18:02rb-bluebotBuild Server message: Build round completed after 957 seconds.
13:18:04rb-bluebotBuild Server message: Revision 49b6853c19 result: All green
13:24:34speachythe idea being a self-contained binary that doesn't need anything other than libc.
13:25:14speachythat decision was made long before my time though, and I'm happy to scrap it since in practice it hasn't worked too well for a while.
13:45:08 Join davisr [0] (~davisr@fsf/emeritus/davisr)
13:45:47***Saving seen data "./dancer.seen"
13:49:44 Join jj5 [0] (~jj5@100.80.216.139.dynamic.dsl.dv.iprimus.net.au)
14:00
14:13:21tertuthanks speachy! i'll give it a shot once i can reboot
14:22:27tertuAlright, well it has already gotten further than it did before
14:31:35tertuhuh apparently the eros ten is based on the x1000 too
14:31:44tertuor well, at least supposedly
14:32:50speachyyeah, I have a rebadged version of one of those lying around here
14:33:23tertui can't tell if the size seems nice or terrible
14:33:29tertui guess it depends on how chunky it is
14:33:43speachyit's decent for what it is, but there's no getting around the 128x64 resolution.
14:34:16speachymy beef are the very limited number of buttons
14:36:53sporkis the volume knob a button ?
14:37:39speachyno, just a knob.
14:38:33speachyfive buttons plus the wheel.
14:38:37tertuyeah I think a friend had a rockboxed clip and it was... okay
14:38:54tertuway back when
14:39:02sporkclip+ i use daily with rockbox
14:39:04tertui have an eros q now and i am perfectly happy with it
14:39:11sporkbut it has many buttons
14:39:16tertutrue
14:39:46tertuoh actually i think it's an eros k but that doesn't really matter
14:39:47speachythe clip's screen is _tiny_.
14:40:00sporksame resolution though
14:40:12speachythe Eros TEN's screen is about the size of the Clip's. Matbe slightly larger
14:40:47speachythe xDuoo X3 has the same resolution in a noticably larger screen.
14:41:09sporkwhat is the name of the rebadged one you have ?
14:41:26speachyHmusic something
14:42:18sporkP68
14:46:28sporkbit odd to have such a limited screen on a fairly expensive device
14:57:59speachyI paid $40 on ebay
15:00
15:04:21sporkfor that kind of money it could be a clip+ successor
15:04:44sporkwith too few buttons
15:06:07speachyit's 2x the size of a clip, and with less utility. and was never produced in any meaningful quantity.
15:06:36speachythe clip series is the counter example to "impossible to put good controls on a small device" complaints.
15:11:12sporkthe world changed in the 15-20 years since the sansa's were released
15:11:32sporkthat makes that eros ten a weird product
15:11:54sporkthe market seems to want touchscreens and streaming
15:12:18sporkor basically just smartphones
15:13:41speachythe eros 10 was released (and discontinued) what, 6-8 years ago?
15:13:54 Quit Natch (Remote host closed the connection)
15:14:39speachythere's still a lot of experimentation with DAPs out of China, but most of the market is either "crappy smartphone minus the phone part" and ultra-low-end shovelware.
15:15:43speachybut they're still actively making/updating the 6-year-old erosq/k family, so kudos
15:19:26sporki hope my sansa's will outlive me
15:21:09speachyI feel that way about the xDuoo X3. I have a spare, and another one that won't power on that I purchased for parts/repair.
15:36:18 Quit jacobk (Ping timeout: 248 seconds)
15:37:16sporki missed the window when that one was sold, it looks very capable
15:37:45 Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net)
15:45:48***Saving seen data "./dancer.seen"
15:58:49 Join Natch [0] (~natch@c-9e07225c.038-60-73746f7.bbcust.telenor.se)
16:00
16:25:10 Quit othello7 (Quit: othello7)
16:30:03tertui found a working clip+ with a dead battery at my local ewaste place recently... hopefully somebody thought to keep the thing
16:30:10 Quit jj5 (Ping timeout: 248 seconds)
16:30:19tertu(i ran out of time to investigate it more fully)
16:32:18 Join jj5 [0] (~jj5@100.80.216.139.dynamic.dsl.dv.iprimus.net.au)
16:38:10 Quit jj5 (Ping timeout: 248 seconds)
16:46:32 Quit speachy (Quit: WeeChat 4.4.2)
16:50:42 Join speachy [0] (~speachy@rockbox/developer/speachy)
16:50:42Mode"#rockbox +v speachy" by ChanServ (ChanServ@services.libera.chat)
16:58:19 Quit advcomp2019 (Ping timeout: 264 seconds)
16:59:35 Join advcomp2019 [0] (~advcomp20@user/advcomp2019)
17:00
17:00:42 Join Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net)
17:06:47 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
17:09:34 Quit advcomp2019 (Ping timeout: 260 seconds)
17:41:48 Join advcomp2019 [0] (~advcomp20@user/advcomp2019)
17:43:59 Quit advcomp2019_ (Ping timeout: 260 seconds)
17:45:51***Saving seen data "./dancer.seen"
17:47:20 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
17:50:46 Quit advcomp2019 (Ping timeout: 252 seconds)
17:57:50 Quit Bubblegumdrop (Quit: ZNC 1.9.1+deb1 - https://znc.in)
18:00
18:00:27 Join Bubblegumdrop [0] (~znc@user/Bubblegumdrop)
18:04:22tertualright all is working very well, thank you everyone
18:06:35 Quit Nezumi-sama (Ping timeout: 252 seconds)
18:13:44 Quit advcomp2019_ (Ping timeout: 260 seconds)
18:14:10speachyanyone out there willing to be a guinea pig for some experimental test builds with m68k (iriver h1xx/3xx) or ipod6g, let me know.
18:16:53speachyDecided to uprev the GCC 8.5.0 toolchain bump to 9.5.0 due to an arm codegen issue that might cause some issues.
18:18:25 Quit cstine (Quit: The Lounge - https://thelounge.chat)
18:18:55 Join cstine [0] (~cstine@150.136.136.191)
18:32:53 Quit buZz (Ping timeout: 248 seconds)
18:34:56 Join buZz [0] (~buzz@drivesthe.skoolbuzz.net)
18:35:20 Nick buZz is now known as Guest4780 (~buzz@drivesthe.skoolbuzz.net)
18:37:42 Join advcomp2019 [0] (~advcomp20@user/advcomp2019)
18:44:28 Join Nezumi-sama [0] (~narf@syn-067-053-148-069.biz.spectrum.com)
18:50:28 Join massiveH [0] (~massiveH@2600:4040:a982:dc00:8d1a:e6d1:d7c6:4933)
19:00
19:09:48 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
19:13:18 Quit advcomp2019 (Ping timeout: 276 seconds)
19:16:47 Join advcomp2019 [0] (~advcomp20@user/advcomp2019)
19:16:59 Quit advcomp2019_ (Ping timeout: 252 seconds)
19:23:06 Nick Guest4780 is now known as buZz (~buzz@drivesthe.skoolbuzz.net)
19:28:18 Quit davisr (Remote host closed the connection)
19:33:43 Quit Natch (Ping timeout: 264 seconds)
19:37:23 Join Natch [0] (~natch@c-9e07225c.038-60-73746f7.bbcust.telenor.se)
19:41:15 Quit advcomp2019 (Ping timeout: 246 seconds)
19:45:55***Saving seen data "./dancer.seen"
19:46:02 Join advcomp2019 [0] (~advcomp20@user/advcomp2019)
19:51:14 Quit jacobk (Ping timeout: 248 seconds)
20:00
20:15:29 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
20:17:08 Join advcomp2019__ [0] (~advcomp20@user/advcomp2019)
20:18:18 Quit advcomp2019 (Ping timeout: 276 seconds)
20:20:21 Quit advcomp2019_ (Ping timeout: 252 seconds)
21:00
21:04:38braewoods_speachy, i've seen a number of handhelds that are running custom Linux firmwares. It's kinda interesting but they're not really suited for rockbox in terms of overall form factor.
21:05:39braewoods_But I was like, a retro handheld has 1GB of RAM? that's a lot considering how much you usually have to work with.
21:16:26 Quit Trzyzet (Quit: The Lounge - https://thelounge.chat)
21:18:48 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
21:18:50 Quit Trzyzet (Client Quit)
21:18:54 Join retr0id1 [0] (~Retr0id@user/retr0id)
21:20:21 Quit retr0id (Read error: Connection reset by peer)
21:20:21 Nick retr0id1 is now known as retr0id (~Retr0id@user/retr0id)
21:27:38 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
21:31:21 Quit Trzyzet (Client Quit)
21:31:38 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
21:39:34 Quit advcomp2019__ (Ping timeout: 252 seconds)
21:45:59***Saving seen data "./dancer.seen"
21:50:43 Quit Trzyzet (Quit: The Lounge - https://thelounge.chat)
21:50:59 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
21:59:06 Quit Trzyzet (Quit: The Lounge - https://thelounge.chat)
22:00
22:00:40 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
22:10:11 Quit Trzyzet (Quit: The Lounge - https://thelounge.chat)
22:10:27 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
22:21:37 Join advcomp2019 [0] (~advcomp20@user/advcomp2019)
22:24:31 Quit Trzyzet (Quit: The Lounge - https://thelounge.chat)
22:24:45 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
22:25:33 Quit Trzyzet (Client Quit)
22:25:49 Join Trzyzet [0] (~Trzyzet@88.97.248.229)
22:52:49 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
22:56:15 Quit advcomp2019 (Ping timeout: 276 seconds)
22:57:00rb-bluebotBuild Server message: New build round started. Revision 7d60b13584, 345 builds, 10 clients.
22:57:00rb-bluebotpdf50606: Fix questionable code that GCC9 complains about by Solomon Peachy
23:00
23:12:37rb-bluebotBuild Server message: Build round completed after 938 seconds.
23:12:39rb-bluebotBuild Server message: Revision 7d60b13584 result: All green
23:12:46rb-bluebotBuild Server message: New build round started. Revision 8b9ef7cf8d, 345 builds, 9 clients.
23:12:47rb-bluebotsdl: Remove SDL_RENDERER_ACCELERATED flag by Christian Soffke
23:26:35rb-bluebotBuild Server message: Build round completed after 829 seconds.
23:26:36rb-bluebotBuild Server message: Revision 8b9ef7cf8d result: All green
23:27:49 Join advcomp2019__ [0] (~advcomp20@user/advcomp2019)
23:30:57 Quit advcomp2019_ (Ping timeout: 244 seconds)
23:46:03***Saving seen data "./dancer.seen"

Previous day | Next day