| 01:00 |
| 01:21:46 | *** | Saving seen data "./dancer.seen" |
| 02:00 |
| 02:11:37 | | Quit othello7 (Quit: othello7) |
| 03:00 |
| 03:21:48 | *** | Saving seen data "./dancer.seen" |
| 04:00 |
| 04:20:30 | | Quit PheralSparky (Read error: Connection reset by peer) |
| 05:00 |
| 05:04:24 | | Quit jacobk (Ping timeout: 260 seconds) |
| 05:05:21 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 05:21:50 | *** | Saving seen data "./dancer.seen" |
| 07:00 |
| 07:21:52 | *** | No seen item changed, no save performed. |
| 09:00 |
| 09:21:53 | *** | No seen item changed, no save performed. |
| 09:34:40 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 10:00 |
| 10:35:31 | | Quit dconrad (Remote host closed the connection) |
| 11:00 |
| 11:15:11 | | Join Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net) |
| 11:21:54 | *** | Saving seen data "./dancer.seen" |
| 11:58:58 | | Join steelbas [0] (~steelbas@185.65.134.180) |
| 12:00 |
| 12:10:15 | _bilgus | per gerrit If a redirect file exists then any of the newer multiboot fw is going to pick this up and use it regardless of where you actually loaded it from |
| 12:10:15 | _bilgus | the problem with loading the firmware with the bootloader is that once the fw is in ram all origin is lost |
| 12:10:15 | _bilgus | in the bootloader multiboot uses bootdata to signal what drive/partition it used |
| 12:10:15 | DBUG | Enqueued KICK _bilgus |
| 12:10:15 | _bilgus | and reads the redirect file but tells the fw nothing about the end status |
| 12:10:15 | _bilgus | in theory if the bootloader couldn't read it then the fw can't either |
| 12:10:16 | *** | Alert Mode level 1 |
| 12:10:16 | _bilgus | so if anything we need to check (in fw) if the redirect file actually points to a real directory with .rockbox/validfw.bin |
| 12:10:19 | _bilgus | mind you I think some form of this verification already exists |
| 12:12:54 | _bilgus | ok so we have verify_boot_data and I think we just need to do a check like the bootloader does to make it bulletproof or just verify that the bin exists at the very least |
| 12:19:19 | speachy | ....extend bootdata to include this info |
| 12:19:21 | speachy | ? |
| 12:20:17 | *** | Alert Mode OFF |
| 12:25:32 | _bilgus | I think I can just check it from the fw |
| 12:26:13 | _bilgus | I have 3 places I can do it so far |
| 12:28:14 | _bilgus | I can do it on the actual redirect volume_onmount_internal I could do it in verify_boot_data or I could do it somewhere bit lower |
| 12:29:57 | _bilgus | I'm thinking we just verify a non empty bin file on the other end |
| 12:30:48 | _bilgus | but we could always extend that to checking for a valid file too but I don't think we do that except in the bootloader |
| 12:53:40 | | Join nift4 [0] (~nick@132.145.243.111) |
| 12:55:53 | nift4 | Hi, is pamaury still active around here? I'm interested in working on LCD sleep for the NWZ-E370/E380 series port and was wondering if pamaury perhaps has some pointers for me. |
| 12:56:58 | nift4 | If not, I'll just try whatever myself, but figured I might as well ask. |
| 13:00 |
| 13:03:25 | _bilgus | unfortunately no I haven't seen pamaury in at least 7 years |
| 13:07:20 | nift4 | okay, thanks for the answer :) I'll just play with it and compare with the linux source code whenever I get a chance |
| 13:12:08 | nift4 | I do have one more question already, what kernel does rockbox use? Is it a custom kernel, or based on some microkernel? Because the source of the e370 port does not seem to be using vendor's linux kernel. On the other hand I saw an Android port which would suggest using an external kernel. |
| 13:14:31 | _bilgus | dconrad g#5887 I think is about the best we (I) can do |
| 13:14:34 | rb-bluebot | Gerrit review #5887 at https://gerrit.rockbox.org/r/c/rockbox/+/5887 : Add verification to multiboot redirect by William Wilgus |
| 13:15:21 | _bilgus | nift4, we do everything ourselves in most cases if its a hosted port then we run ontop of the existing linux |
| 13:16:03 | _bilgus | I'm not familiar with the sonys I know they are a pain |
| 13:17:01 | nift4 | thanks, that clears things up |
| 13:17:03 | _bilgus | pretty sure pamaury only did native ports but you'd probably need to look at the wiki for your specific player |
| 13:17:43 | _bilgus | https://www.rockbox.org/wiki/SonyNWZE370Port.html |
| 13:18:41 | _bilgus | I don't know how much I'd trust that zip in there but it atleast tells me its a native port |
| 13:18:57 | _bilgus | so no kernel its all us |
| 13:21:56 | _bilgus | 240816 Aug 24th 2016 yeah thats a pretty old version but at least you have a way to test a known good one |
| 13:21:58 | *** | No seen item changed, no save performed. |
| 13:22:30 | _bilgus | because really we might have supported these players in 2016 but if no one has tested them WHO KNOWS |
| 13:25:12 | nift4 | why is the latest zip from 2016? do these not-really-supported devices not get any development builds? |
| 13:28:41 | nift4 | oh and while at it, https://www.rockbox.org/wiki/ReleaseNotes315 says these sony players are "stable" and I can see they indeed got a 3.15 stable release - but then the devices page doesn't list them as stable. Any idea what might have happened here? |
| 13:40:16 | _bilgus | like I said that was a known working build I don't have the device and I can't really say what state you'll find it in other than should work |
| 13:41:24 | _bilgus | we have current dev builds for it.. |
| 13:43:11 | nift4 | oh where would i find the current builds |
| 13:43:29 | _bilgus | https://www.rockbox.org/daily.shtml |
| 13:44:11 | | Quit rb-bluebot (Remote host closed the connection) |
| 13:44:19 | | Join rb-bluebot [0] (~rb-bluebo@rockbox/bot/utility) |
| 13:44:23 | _bilgus | and if a player is supported you should be able to use https://www.rockbox.org/download/ rbutil to get 3.15 or the latest dev on there |
| 13:45:01 | nift4 | i did try rbutil but even though there are 3.15 builds the device just isn't in the list in rbutil (unless i'm a moron but I triple checked theres no sony, not even in hidden) |
| 13:45:19 | _bilgus | no that probably means manual install |
| 13:45:23 | nift4 | so I followed the ImxInstall guide |
| 13:45:48 | nift4 | oh I see what happened |
| 13:45:51 | nift4 | the battery is dead |
| 13:45:57 | nift4 | it discharged during the update |
| 13:46:06 | _bilgus | thats scary |
| 13:46:06 | nift4 | I should've left it plugged, I'm a genius lmao |
| 13:47:02 | nift4 | well I mean I am in the OF bootloader USB mode and my host can see a [Sigmatel, Inc. ROM Recovery], but I suppose I need to search for fixes elsewhere lol |
| 13:47:04 | _bilgus | well that install guide shoud get you a bootlader then you can use it to load 3.15 or (preferably the dev version) |
| 13:47:21 | nift4 | atleast I have a backup of the OF downloaded from that wiki page |
| 13:47:36 | _bilgus | No for sure you have a way back from that |
| 13:48:02 | _bilgus | I think there is a recovery tool in or tree |
| 13:48:28 | _bilgus | the wiki should have some explanation |
| 13:48:34 | nift4 | will check thx |
| 13:49:54 | _bilgus | https://www.rockbox.org/wiki/ImxInstall.html#Expert_instructions |
| 13:50:29 | _bilgus | look where it says building a recovery img |
| 13:50:59 | nift4 | the USB recovery mode hmm, I'm not yet quite sure how I can use it (or if I can on linux at all) |
| 13:51:32 | _bilgus | i've done it when pamaury was here so yes it works but I can't tell you anything that sticks out in my mind about it though |
| 13:51:47 | _bilgus | 7+ years ago |
| 13:53:02 | _bilgus | may have some luck in the irc logs |
| 13:55:00 | nift4 | should I attempt to build a recovery image and use sbloader? https://www.rockbox.org/wiki/STMP37xxRecovery.html |
| 13:56:07 | _bilgus | https://www.rockbox.org/irc/log-20160922 |
| 13:56:12 | _bilgus | around 22:50 |
| 13:56:49 | _bilgus | https://www.rockbox.org/irc/log-20160922#22:49:28 |
| 13:57:06 | _bilgus | didn't realize it did that |
| 13:57:54 | nift4 | thanks alot for your help, will read through that. just to clarify firmware.sb here refers to dump of OF? |
| 13:59:21 | _bilgus | uh let me look because any file named that is usually what gets loaded |
| 13:59:49 | nift4 | sudo dd if=firmware.sb bs=512 seek=4 of=/dev/sdb2 |
| 14:00 |
| 14:00:08 | nift4 | was talking about this part |
| 14:00:09 | _bilgus | of/firmware.sb is for sure the OF one |
| 14:01:07 | _bilgus | so the I would say its the OF one if you want to restore or its the bootloader one if you want rb too |
| 14:01:26 | _bilgus | should be perfectly safe to supply the backed up OF one either way |
| 14:01:44 | nift4 | i suppose i'll try to get back clean OF first and then try to install rb the proper way again after that |
| 14:01:59 | nift4 | so.. time to get to work :D |
| 14:02:08 | _bilgus | yeah being that you'll know after that makes sense to me |
| 14:03:00 | _bilgus | any hey maybe you'll catch the bug and come be the new Sony expert round here? |
| 14:03:10 | _bilgus | "p no pressure |
| 14:03:18 | _bilgus | :p |
| 14:03:27 | nift4 | lol |
| 14:04:48 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 14:05:34 | nift4 | wait a minute nevermind, I am actually seeing a "Rockbox media player |
| 14:05:47 | nift4 | in my usb, while screen still says Bootloader USB mode |
| 14:05:59 | nift4 | so I just booted the rb bootloader and scared myself to hell? -_- |
| 14:07:02 | nift4 | o h |
| 14:07:05 | nift4 | it works |
| 14:07:22 | nift4 | and i just pressed one button to many and prevented myself from booting up successfully. lol |
| 14:09:28 | nift4 | ok yeah big facepalm. but hey i have working rb now |
| 14:11:51 | | Join othello7 [0] (~Thunderbi@pool-100-36-176-164.washdc.fios.verizon.net) |
| 14:14:31 | nift4 | no i dont think its working properly |
| 14:14:49 | nift4 | it only boots if i go to ROM Recovery mode and then unplug usb cable, then it decides to start rb |
| 14:15:39 | nift4 | if I just turn it off and try to turn it on, I get a blackscreen and nothing happens |
| 14:18:11 | nift4 | i unplug usb, turn off rockbox, hold the power button but nothing happens. if I instead turn it off, plug in usb cable, I get to bnootloader usb mode, then unplug and I get to rockbox |
| 14:18:13 | | Quit steelbas (Quit: Leaving) |
| 14:19:11 | nift4 | I wonder why it doesn't boot without cable, is the battery just that dead or did I semi brick it? confused |
| 14:26:48 | _bilgus | we do have stuff that stops boot if the battery is too low but not sure TBH |
| 14:44:39 | | Quit dconrad (Remote host closed the connection) |
| 15:00 |
| 15:00:55 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 15:06:30 | | Quit dconrad (Remote host closed the connection) |
| 15:13:26 | nift4 | _bilgus: can you show me roughly where the code for this is? I did see battery_level_safe() but it doesn't seem to be responsible for that |
| 15:16:14 | nift4 | ah i think i found it: query_force_shutdown() |
| 15:22:01 | *** | Saving seen data "./dancer.seen" |
| 15:28:25 | | Join sam_d [0] (~sam@user/sam-d/x-8933526) |
| 15:34:05 | _bilgus | yeah you can just define NO_LOW_BATTERY_SHUTDOWN or make it #if 1 |
| 15:35:05 | _bilgus | I'll bbl feel free to ask others lurking might be able to help too :) |
| 15:36:21 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 15:42:35 | dconrad | _bilgus: thanks for 5887, I'll take a look see |
| 15:43:30 | dconrad | I may need to do some dissecting, because clearly it worked some time ago |
| 15:46:24 | _bilgus | was that before multiboot was extended to the device perhaps or before v2 where its pushing the drive and partition though? |
| 15:47:30 | _bilgus | also if the refdirect is present it should have always prefered to use that anyway so maybe I'm confused |
| 15:48:15 | _bilgus | I'm on my way out but i'll check my phone in a while |
| 15:49:46 | dconrad | Sounds good. To be clear, when the redirect doesn't work and the root install /.rockbox is not present, it simply states that it couldn't find rockbox |
| 15:50:19 | dconrad | I'll see if I can narrow it doewn |
| 15:50:21 | dconrad | down* |
| 15:52:57 | | Quit jacobk (Ping timeout: 276 seconds) |
| 15:53:21 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 16:00 |
| 16:08:18 | | Quit jacobk (Ping timeout: 248 seconds) |
| 16:08:56 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 16:10:09 | dconrad | AHA |
| 16:10:28 | dconrad | https://git.rockbox.org/cgit/rockbox.git/commit/?id=dc364e44ef6aeef4082303bc07300d85c5852ba0 |
| 16:10:55 | dconrad | I think we need to separate out some stuff from HAVE_MULTIVOLUME, or something to have redirect work in bootloaders |
| 16:11:42 | dconrad | I don't rightly know if we can separate out the necessary HAVE_MULTIVOLUME functionality to a different define flag though |
| 16:15:59 | dconrad | So is the intended way to do redirect still that bootloader reads redirect file and goes directly to the redirect directory? |
| 16:16:44 | dconrad | Or is it something like the bootloader always loads the standard /.rockbox, then that redirects itself to the redirect directory? That would seem weird |
| 16:17:12 | dconrad | I guess I was assuming the former is how it's intended to work |
| 16:26:18 | dconrad | I can't seem to find that patchset in gerrit, I presume it was pushed directly to git |
| 16:32:13 | | Quit dconrad (Remote host closed the connection) |
| 16:56:44 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 17:00 |
| 17:22:05 | *** | Saving seen data "./dancer.seen" |
| 17:50:37 | dconrad | I put up g#5888, which just adds HAVE_MULTIVOLUME back into the bootloaders, that works, though I'm sure the bin size is going to be bigger than it needs to be |
| 17:50:40 | rb-bluebot | Gerrit review #5888 at https://gerrit.rockbox.org/r/c/rockbox/+/5888 : Revert "Disable HAVE_MULTIVOLUME for bootloaders" by Dana Conrad |
| 17:58:50 | nift4 | so in the end, I rebuilt bootloader (and successfully flashed, also changed some text to verify that) and rockbox image with NO_LOW_BATTERY_SHUTDOWN but it still won't boot rockbox. I can boot OF (by pressing back) without cable, or I boot either rockbox or OF by plugging and unplugging cable, but I can't boot rockbox without cable. Going to sleep now |
| 17:59:27 | nift4 | but hey atleast latest builds seem to have no regressions, both bootloader and rockbox |
| 18:00 |
| 18:11:33 | | Quit nift4 (Quit: leaving) |
| 18:57:27 | | Quit dconrad (Remote host closed the connection) |
| 19:00 |
| 19:22:06 | *** | Saving seen data "./dancer.seen" |
| 20:00 |
| 20:14:50 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 20:20:12 | | Join PheralSparky [0] (~S|h|a|w|n@user/shawn/x-4432647) |
| 21:00 |
| 21:02:39 | | Quit Natch (Remote host closed the connection) |
| 21:05:54 | | Quit dconrad (Remote host closed the connection) |
| 21:11:50 | | Join Natch [0] (~natch@c-9e07225c.038-60-73746f7.bbcust.telenor.se) |
| 21:15:50 | __builtin | Can someone give g#5889 a test? There should be two new puzzles, sgt-slide and sgt-sokoban. |
| 21:15:53 | rb-bluebot | Gerrit review #5889 at https://gerrit.rockbox.org/r/c/rockbox/+/5889 : puzzles: add Slide and Sokoban. by Franklin Wei |
| 21:16:31 | __builtin | I suspect sgt-slide will sporadically crash due to memory exhaustion. |
| 21:18:42 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 21:21:00 | _bilgus | dconrad that sounds like just a bad flag somewhere good find! |
| 21:22:11 | *** | Saving seen data "./dancer.seen" |
| 21:22:57 | _bilgus | here perhaps https://github.com/Rockbox/rockbox/blob/master/firmware/export/config.h#L908 |
| 21:25:02 | _bilgus | so I guess it can get changed to drop mv #if defined(HAVE_BOOTDATA) && defined(BOOT_REDIR) |
| 21:28:29 | _bilgus | __builtin, /plugins/plugin_crt0.o: In function `plugin__start': |
| 21:28:29 | _bilgus | plugin_crt0.c:(.text+0x60): undefined reference to `plugin_start' |
| 21:28:29 | _bilgus | collect2: error: ld returned 1 exit status |
| 21:28:29 | _bilgus | plugins/plugins.make:163: recipe for target 'plugins/puzzles/sgt-slide.rock' failed |
| 21:28:49 | _bilgus | I did make clean first |
| 21:29:23 | _bilgus | sgt-sokoban.rock saya the same |
| 21:50:13 | | Quit Moriar (Quit: Leaving.) |
| 22:00 |
| 22:27:06 | _bilgus | https://www.theverge.com/2024/8/16/24221635/microsoft-fat32-partition-size-limit-windows-11 |
| 22:27:32 | _bilgus | only the commandline though |
| 22:45:49 | | Quit braewoods (Remote host closed the connection) |
| 22:46:03 | | Join braewoods [0] (~braewoods@user/braewoods) |
| 23:00 |
| 23:22:14 | *** | Saving seen data "./dancer.seen" |
icon identifies that the person is a core developer (has commit access).