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 2021-04-17

00:10:44***Saving seen data "./dancer.seen"
00:26:54braewoodsRetr0id: afaik, no. i think there hasn't been much interest in messing with it because we can just get bootstrapped from the HDD ala BIOS style
00:27:37braewoodsRetr0id: afaik the only ports where we do that currently are H100, H120, H300 ports
00:27:48braewoodsand they're not even PP
00:27:51braewoodsColdFire based
00:28:42braewoodshaving to update the ROM chip to perform a bootloader update is more risky also so that's probably another reason we don't
00:28:52braewoodsexcept when absolutely required
00:29:43braewoodsi took that risk when producing new bootloaders for those ports. I thought we should have a new bootloader after 10 years because of some outstanding bugs they had.
00:37:18 Quit ac_laptop (Ping timeout: 240 seconds)
00:37:22 Quit braewoods (Quit: WeeChat 2.8)
00:42:36 Join braewoods [0] (~braewoods@learnprogramming/staff/braewoods)
01:35:27 Join flyback [0] (~flyback@2601:540:8201:1e60:71c5:1baa:f9e2:db33)
01:35:51flybackwell I actually have a m200v4 lying around (finding it tomarrow) if anyone wants me to test code
01:45:52flybackdidn't think anyone worked on that port anymore but I see a bunch of stuff has changed
02:10:47***Saving seen data "./dancer.seen"
02:43:21 Quit icypee (Quit: Connection closed)
03:21:33 Join ZincAlloy [0] (
03:32:13 Quit pixelma (Ping timeout: 250 seconds)
03:32:13 Quit amiconn (Ping timeout: 250 seconds)
03:32:45 Join pixelma [0] (marianne@rockbox/staff/pixelma)
03:33:08 Join amiconn [0] (jens@rockbox/developer/amiconn)
03:38:29 Join lebellium [0] (
03:42:28 Join ufdm_ [0] (
03:42:37 Quit ufdm (Read error: Connection reset by peer)
03:43:30 Quit ufdm_ (Client Quit)
03:43:44 Join ufdm [0] (
04:10:49***Saving seen data "./dancer.seen"
06:10:53***No seen item changed, no save performed.
06:28:32 Join MrZeus_ [0] (~MrZeus@2a02:c7f:a0aa:4400:ddab:dec4:15ba:9941)
06:47:57braewoodsflyback: it appears the port is mostly finished but isn't marked stable for some reason
06:48:01braewoodsno explanation is given
06:48:13braewoodsmaybe there's one in the commit log
07:02:36 Quit michaelni (Ping timeout: 265 seconds)
07:11:36Retr0idbraewoods, understandable, I've been taking the risk because the iPod nano 1g is only like £10 on ebay these days, so I can afford to brick a few.
07:15:19braewoodsRetr0id: you're welcome to try. you should try to understand the boot process before attempting it though.
07:15:32braewoodsCF is totally different from PP
07:15:37braewoodsthough some similarities
07:15:49 Join michaelni [0] (
07:16:06braewoodson CF we change the reset vector in the eeprom to boot our payload that we store near the end of the flash
07:16:25braewoodsit's the last 64K of the EEPROM that the OF doesn't use
07:17:18braewoodsRetr0id: and if the EEPROM is one of those VFS chips i think it wa
07:17:57braewoodsRetr0id: you can look at how we program those in iriver_flash. it's a port specific plugin for doing various flashing operations on the EEPROMs in the H100 and H300 series
07:18:37braewoodsi refactored it to allow it work on both variants
07:19:06Retr0idyeah, I currently have a hacked up version of iriver_flash
07:19:09braewoodsi don't know how PP boots exactly but i suspect it may have a similar set of interrupt vectors if the EEPROM is stored at address 0
07:19:32braewoodsexcept on ARM, it's not addresses.
07:19:43braewoodsit's an array of instructions.
07:19:47braewoodssingle instructions.
07:19:55braewoodsthey jump to where the actual code is stored
07:20:15braewoodsnot sure if that's relevant to the EEPROM bootloader but it's used in the rockbox and OF firmware payloads
07:20:29Retr0idthe ipod ROM has a "partition table" type thing, much like the one at the start if the HDD's firmware partition. It stores the "diag" and "disk" mode boot images. Rather than attempt to modify the actual bootloader, I just added a new entry to the table
07:20:44braewoodsi see
07:20:52braewoodsCF has no such system in their eeproms
07:20:57braewoodsit's just raw payloads
07:21:40braewoodsonly the interrupt vectors are special afaik
07:22:34braewoodsi have no idea if it'll work but maybe you can embed the existing bootloader minus the usual mi4 container
07:22:54braewoodsin iriver_flash we add a small header for safety reasons
07:22:57braewoods8 bytes, iirc
07:23:26Retr0idOne weird thing that took me a long time to figure out, is that apple wired up the ROM address lines in the wrong order, so the addresses you need to write to, to enable programming mode etc. are different
07:23:50*braewoods shrugs.
07:23:55braewoodsit may have been intentional
07:24:19braewoodsanother obfuscation technique
07:24:30braewoodsit could also be differences in endianness
07:24:34braewoodsCF is big endian
07:24:42braewoodsARM is typically little endian
07:24:49Retr0idnah it's not just endianness, it's a bitwise order reversal
07:24:55Retr0idmy theory is that it made the pcb routing easier
07:24:56braewoodsI see.
07:25:18braewoodsin any case expect this to be very device specific
07:25:34braewoodsother PP targets probably can't do this without device specific hackery
07:25:36Retr0idI *believe* it's consistent between all portalplayer ipods
07:25:46braewoodsah i meant PPs that aren't ipods
07:26:07braewoodssome of my PPs are so rare i wouldn't even attempt this
07:26:40braewoodsthen again the OF for ipods is a lot different than for general PP
07:27:13Retr0idOF = original firmware?
07:27:37braewoodsiirc, apple insists on their own custom disk partitioning and other crap
07:27:49braewoodsthe other PPs are generally more flexible and laid back on such things
07:27:58braewoodsor is that just later ipods?
07:28:27braewoodsi can reformat my iriver ipod with linux tools quite easily
07:28:35braewoodsiriver h10
07:28:42Retr0idpp ipods are generally pretty flexible. The firmware needs to be in the first partition, but everything else is flexible
07:28:48braewoodsi see
07:28:50Retr0idI've reformatted with linux tools
07:29:00braewoodsi guess that's more of a later problem
07:29:10braewoodsi just don't see the point of putting our bootloader in ROM
07:29:22braewoodswhen we have no problems with the existing bootloader
07:29:24Retr0idyeahhhh it is more of a "because I can" thing
07:29:42braewoodson CF it kinda made sense to do that
07:30:01braewoodsthe OF ran entirely from ROM and provided no method to boot from disk
07:30:17braewoodsso we had to embed our bootloader there anyway
07:30:32braewoodsno early bootloader for us to leverage
07:31:14braewoodsPP is designed to load the actual OF from disk like a PC bios
07:31:40braewoodsso not much need to hack the ROM bootloader
07:31:45braewoodseasier to just work with it :)
07:36:50 Quit michaelni (Ping timeout: 252 seconds)
07:50:12 Join michaelni [0] (
07:53:50braewoodsinteresting. gigabeat S had one last OF release of 1.3
07:53:53braewoodsI wonder...
08:05:23speachyicypee: AFAIK the mini1g is supposed to be stable, but it's not something I can recall anyone talking about for a while. I know hte mini2g generally just works. Key question −− have you modded your mini1g in any way, especially replacing the storage?
08:07:39braewoodsspeachy: wrong person?
08:07:51braewoodsI thought it was Retr0id that asked about it.
08:08:00braewoodsI see.
08:08:08speachybraewoods: that was nano1g
08:08:22braewoodsyea, wasn't paying attention
08:08:30braewoodsso many ipods with small differences
08:09:16braewoodsanywho i'm going to test installing rockbox to this vanilla gigabeat S later
08:09:21fs-bluebotBuild Server message: New build round started. Revision 4dc602dd7f, 298 builds, 10 clients.
08:09:45 Quit michaelni (Ping timeout: 260 seconds)
08:09:48braewoodsi need to understand the manual process before i attempt work on rbutil
08:10:39speachyd'oh, I should not have done that
08:10:54***Saving seen data "./dancer.seen"
08:13:15speachyrbutil wraps the standalone tools
08:13:33 Join michaelni [0] (
08:17:31braewoodsspeachy: i know, this one has one major difference though
08:17:40braewoodsit has to use MTP to upload the initial payload
08:17:54braewoodsor so i was told
08:18:20braewoodsit would be nice to finally move gigabeat s to stable
08:18:25braewoodsthe port doesn't need much work
08:18:44braewoodsin any case one thing at a itme
08:22:08fs-bluebotBuild Server message: Build round completed after 768 seconds.
08:22:11fs-bluebotBuild Server message: Revision 4dc602dd7f result: All green
08:36:55speachy_bilgus: so is g#3330 ok to merge?
08:36:58fs-bluebotGerrit review #3330 at : Fix Pictureflow bugs by Aidan MacDonald
08:47:18speachydconrad, see my post-merge comment on 3228.
09:06:13_bilgusspeachy 3330 is good
09:06:31fs-bluebotBuild Server message: New build round started. Revision afe80742a5, 298 builds, 10 clients.
09:06:48_bilgusdid you see g#3332
09:06:50fs-bluebotGerrit review #3332 at : 6 bit lcd_mono_bitmap by William Wilgus
09:07:03speachy(the joys of quiet dreary rainy morning hack time while the household is still asleep)
09:08:15speachy3332 isn't clear from teh description; is it an out-of-bounds read, or write?
09:09:04speachyI mean, it's a bug either way but if it's just a read then it won't lead to corruption
09:12:27 Join cockroach [0] (~blattodea@pdpc/supporter/active/cockroach)
09:13:08_bilgusyeah just a read
09:13:24_bilgusso wrong but not really wrong
09:18:18fs-bluebotBuild Server message: Build round completed after 706 seconds.
09:18:20fs-bluebotBuild Server message: Revision afe80742a5 result: All green
09:18:32 Join Saijin_Naib [0] (
09:23:26 Quit Saijin_Naib (Ping timeout: 258 seconds)
09:45:05 Join MrZeus__ [0] (~MrZeus@2a02:c7f:a0aa:4400:ddab:dec4:15ba:9941)
09:48:03 Quit MrZeus_ (Ping timeout: 248 seconds)
10:07:03 Join lebellium_ [0] (
10:07:11 Quit lebellium (Ping timeout: 246 seconds)
10:10:57***Saving seen data "./dancer.seen"
10:25:07 Quit J_Darnley (Ping timeout: 265 seconds)
10:28:33 Join ac_laptop [0] (~ac_laptop@
10:33:52 Join Saijin_Naib [0] (
10:37:34 Join amachronic [0] (~weechat@
10:39:30 Join J_Darnley [0] (
10:44:19 Quit Saijin_Naib (Ping timeout: 258 seconds)
11:03:32 Join Saijin_Naib [0] (
11:34:01fs-bluebotBuild Server message: New build round started. Revision 85fbbd9c7f, 298 builds, 10 clients.
11:44:24fs-bluebotBuild Server message: Build round completed after 622 seconds.
11:44:26fs-bluebotBuild Server message: Revision 85fbbd9c7f result: All green
12:03:14 Join f1refly [0] (
12:10:59***Saving seen data "./dancer.seen"
12:58:07 Quit MrZeus__ (Ping timeout: 260 seconds)
13:05:14 Quit Rower ()
13:17:10 Join Rower [0] (
13:20:11 Quit f1refly (Quit: see ya in hell)
13:23:21 Join MrZeus__ [0] (~MrZeus@2a02:c7f:a0aa:4400:ddab:dec4:15ba:9941)
13:29:57flybackwell anyways soon as I find it
13:30:05flybackI will test rockbox on this old m200v4
13:30:10flybackit's going to collect dust otherwise
13:31:10flybackmight play with the android port as well
13:33:52flybackandroid port is actually more useful than people might think
13:34:00flybackturning old phones into a decent modplayer
13:42:11 Quit Saijin_Naib (Ping timeout: 258 seconds)
13:59:54 Join dconrad [0] (~dconrad@
14:11:02***Saving seen data "./dancer.seen"
14:29:59 Quit dconrad ()
14:33:52 Join Saijin_Naib [0] (
14:53:42 Join f1refly [0] (
15:24:18 Quit fs-bluebot (Ping timeout: 265 seconds)
15:24:57 Quit bluebrother (Ping timeout: 246 seconds)
15:29:28speachyamachronic: Welcome to the hallowed group of Rockbox Commiters
15:30:05amachronicyep, thanks! glad to be on board.
15:31:12flybackit's a m250 not a m200
15:31:27*flyback grumbles
15:32:48speachy_bilgus: I also updated the irc nick list so the log viewer shows you as a developer too
15:33:38speachyamachronic: we're not big on process beyond public shaming if you break the build; we're just happy to have contributors.
15:34:16flybackunless m250 is considered a m200v4 series
15:34:49flybackI seem to recall something about one of the digits being the size mabye
15:35:19 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
15:36:48 Quit Saijin_Naib (Ping timeout: 258 seconds)
15:37:43flybackm200v4 AS3525 No No No No No Unusable 4.1.08
15:37:50flybackit's the same thing
15:37:56flybackI can test this for you guys
15:38:01flybackand let you know how it goes
15:39:30speachyflyback: thank you
15:39:37flybacksure :)
15:40:23flybackwish I could bring back my h10 but not spending another $20 on a mechaical hard drive that lasted a month
15:43:02 Quit Rower (Ping timeout: 252 seconds)
15:53:56braewoodsflyback: what model of H10?
15:57:11 Join PeppeNT [0] (
15:58:24PeppeNThi speachy
15:58:30PeppeNTi wrote you on the forum
15:58:43speachyI saw. Haven't had the time to dig into that dump yet
15:58:59PeppeNTno problem mate
15:59:05PeppeNTi'm watching into source code
15:59:26PeppeNTand making build with little modifications
15:59:30PeppeNTbut without results
15:59:47PeppeNTstuck on backlight on with power off
16:02:31speachyhave you looked at firmware/powermgnt.c, the shutdown_hw() function?
16:03:32speachyif you comment out lines 788-789, does it make any difference?
16:04:06 Join Saijin_Naib [0] (
16:05:24PeppeNTyep, i'm working on that file, but not sure if is that function you mentioned. i have to switch on linux (now i'm on windows), because i'm working on source code by linux. I'll be back in 5-10 minutes
16:05:35 Quit PeppeNT (Remote host closed the connection)
16:06:36flyback5GB ONE
16:06:51flybackTHEY USED A non standard cf hd interface so I can't just swap in a cf flash
16:08:39flybackI was going to soder up my old interface but never got a chance and my vision has deteriated and hands shake too much
16:11:03***Saving seen data "./dancer.seen"
16:12:28amachronicspeachy: ok I'll try not to break anything :D
16:12:48 Join PeppeNT [0] (
16:13:59PeppeNTok i'm here. i just edited shutdown_hw(), but added lines after sleep(HZ/4);
16:14:33PeppeNTi tried to call functions backlight_off(); and tried to set timeout to 0 with backlight_set_timeout(0);
16:16:12PeppeNTnow i try to comment lines 788-789
16:16:19PeppeNTand make a build
16:16:57flybackno luck
16:17:03flybacktotally ignored the file, didn't update
16:17:29flybackoh well
16:17:36*flyback tries the unstable rockbox utility
16:30:05flybackoh well mabye another time
16:38:29PeppeNTno results
16:39:10PeppeNTonly the first time i reboot, backlight remains off, but by the 2nd time, the issue reappears
16:39:31flybacktried both the manual patching firmware procedure and enabled the disabled rockbox utility installer and in both cases the player just acts like nothing was loaded oh well
16:39:47flybackI got other things to work on, but I will try again sometime to help debug m200v4 port
16:56:38braewoodsflyback: you could always switch to an H10 20GB :D
17:09:25 Quit amachronic (Ping timeout: 252 seconds)
17:11:13 Join amachronic [0] (~amachroni@
17:20:07PeppeNTspeachy no results
17:20:16PeppeNTissue is still alive
17:36:44 Quit PeppeNT (Quit: KVIrc 5.0.0 Aria
18:00:43 Quit ZincAlloy (Quit: Leaving.)
18:11:04***Saving seen data "./dancer.seen"
18:21:57 Join PimpiN8 [0] (PimpiN8@gateway/vpn/nordvpn/pimpin8)
18:38:41 Quit amachronic (Quit: amachronic)
18:51:06 Quit Strife89 (Quit: No Ping reply in 180 seconds.)
18:52:25 Join Strife89 [0] (
18:52:49 Quit _bilgus (Ping timeout: 252 seconds)
18:55:37 Quit pamaury (Ping timeout: 260 seconds)
18:58:27 Quit lebellium_ (Quit: Leaving)
20:11:07***Saving seen data "./dancer.seen"
20:23:54 Quit PimpiN8 (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:08:07 Quit MrZeus__ (Ping timeout: 260 seconds)
21:13:04 Join massiveH [0] (
21:56:36 Join Strife89|Home [0] (
21:57:24 Quit Strife89 (Ping timeout: 252 seconds)
22:11:09***Saving seen data "./dancer.seen"
22:18:20 Quit cockroach (Quit: leaving)
22:32:15 Join _bilgus [0] (
23:12:58 Quit Saijin_Naib (Ping timeout: 258 seconds)

Previous day | Next day