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 2016-04-20

00:03:48 Join CrashBash-Kun [0] (~CrashBash@unaffiliated/crashbash-kun)
00:29:38 Quit ender` (Quit: In politics, absurdity is not a handicap. -- Napoleon Bonaparte)
01:06:19 Nick [Saint_] is now known as [Saint] (~hayden@rockbox/staff/saint)
01:07:42 Quit bertrik (Read error: Connection reset by peer)
01:24:59 Quit PurlingNayuki (Remote host closed the connection)
01:27:44 Join PurlingNayuki [0] (
01:29:09 Quit girafe (Read error: Connection reset by peer)
01:30:20 Quit Coup_ (Ping timeout: 250 seconds)
01:36:44 Quit pamaury (Ping timeout: 252 seconds)
01:52:25***Saving seen data "./dancer.seen"
02:08:19 Quit ZincAlloy (Quit: Leaving.)
02:08:46 Quit TheSeven (Ping timeout: 268 seconds)
02:09:48 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
02:49:41 Quit krabador (Quit: Take The Time)
03:15:01 Quit Rower (Quit: Hmmm...)
03:18:48 Join Rower [0] (
03:39:12 Join rela_ [0] (
03:41:16 Quit rela (Ping timeout: 268 seconds)
03:50:23 Quit PurlingNayuki (Remote host closed the connection)
03:52:27***Saving seen data "./dancer.seen"
04:08:47 Quit prof_wolfff (Ping timeout: 244 seconds)
04:13:03 Join Darkenvy [0] (
04:17:48 Join PurlingNayuki [0] (
04:23:32 Join prof_wolfff [0] (
05:49:50 Quit FSanches (Quit: Leaving.)
05:52:31***Saving seen data "./dancer.seen"
05:53:47 Quit PurlingNayuki (Remote host closed the connection)
06:02:45 Join PurlingNayuki [0] (
06:03:35 Quit PurlingNayuki (Remote host closed the connection)
06:03:50 Join PurlingNayuki [0] (
06:10:43 Quit zoktar (Ping timeout: 244 seconds)
06:12:22 Quit Marex (Ping timeout: 268 seconds)
06:13:02 Quit gevaerts (Ping timeout: 264 seconds)
06:13:24 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
06:20:54 Join Marex [0] (~Marex@
06:21:49 Join zoktar [0] (
06:21:49 Quit zoktar (Changing host)
06:21:49 Join zoktar [0] (~zoktar@unaffiliated/zoktar)
06:56:21bp0does rockbox aac decoder support he-aac v2?
06:56:31bp0there is some conflicting information
06:57:22bp0if course, he-aacv2 will play as mono he-aac without the parametric stereo of v2, but I mean does it actually decode and use the PS information
07:21:50 Join wodz_ [0] (
07:26:54 Quit Darkenvy (Quit: Textual IRC Client:
07:49:11 Quit CrashBash-Kun (Ping timeout: 268 seconds)
07:52:33***Saving seen data "./dancer.seen"
07:58:30 Quit pixelma (Remote host closed the connection)
07:58:30 Quit amiconn (Remote host closed the connection)
08:00:45 Join pixelma [0] (~pixelma@rockbox/staff/pixelma)
08:00:52 Join amiconn [0] (~amiconn@rockbox/developer/amiconn)
08:25:08 Join ender` [0] (krneki@
08:27:04 Quit ender` (Remote host closed the connection)
08:27:33 Join ender` [0] (
08:54:24 Join petur [0] (~petur@rockbox/developer/petur)
09:23:50 Quit shmibs (Ping timeout: 264 seconds)
09:24:37 Quit TD-Linux (Ping timeout: 250 seconds)
09:24:46 Join shmibs [0] (
09:25:46 Join TD-Linux [0] (~Thomas@about/essy/indecisive/TD-Linux)
09:52:37***Saving seen data "./dancer.seen"
10:29:17 Join metaphysis [0] (~metaphysi@
11:09:25 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
11:24:29 Quit pamaury (Read error: Connection reset by peer)
11:52:43***Saving seen data "./dancer.seen"
12:37:51 Join ZincAlloy [0] (
12:43:53 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
13:09:57 Join FSanches [0] (~felipe@2804:14c:32:8e1e:4c7b:6928:789c:e2ef)
13:27:51 Join krabador [0] (~krabador@unaffiliated/krabador)
13:52:46***No seen item changed, no save performed.
14:17:23 Quit krabador (Quit: Take The Time)
14:32:55 Join krabador [0] (~krabador@unaffiliated/krabador)
14:39:31funmanbp0: to be sure you can test with a simulator
14:57:26 Join Mihail [0] (25d467b0@gateway/web/cgi-irc/
15:02:44MihailWe need very simple patch to boot from sd card: g#1294. For next step I want add booting from sd card by default if it have .rockbox folder. It can be useful when internal storage can be mounted but FS broken. Also it can be useful if someone want update rockbox on player with broken usb.
15:16:02 Quit wodz_ (Ping timeout: 240 seconds)
15:17:47funmanMihail: bootloader looks for .rockbox/ in internal storage and then SD if it's not found?
15:17:47funmanah only if it can't be mounted
15:29:20pamauryMihail: did you find the problem with the sd card ?
15:31:12MihailNo. Are you see my yesterday message: I have no experience with this thing. I upload both builds and patch which cause this problem: Can you look at this?
15:39:25 Quit krabador (Quit: Take The Time)
15:52:47***Saving seen data "./dancer.seen"
15:53:59pamauryyeah I saw the diff but it's completely unrelated, so it will be really hard to debug
15:56:11 Join rela [0] (~x@pdpc/supporter/active/rela)
15:56:27Mihailhow can I help or what should I try? It not alignment problem?
15:58:02 Quit rela_ (Ping timeout: 240 seconds)
16:09:18pamauryI don't know, since it is in rolo load, maybe this extra guys time to some code somewhere to properly finish something. And maybe without the delay, rockbox starts in a weird hardware state.
16:09:38pamaurymaybe you should check if indeed all subsystems properly shutdown when roloing
16:09:43pamauryespecially the storage one
16:09:45pamauryand sd
16:09:52pamauryand dma
16:16:07MihailI try boot bad rockbox build from rolo without this patch (from good build) and still have same problem.
16:17:08pamaurybut with the delay it works ?
16:20:46MihailNo. Good build have no delay. It don't work only if new (which I try load through rolo) build have delay.
16:21:24Mihailand only if it on sd card
16:23:10 Quit PurlingNayuki (Quit: PurlingNayuki)
16:24:44pamaurywait but it does not make sense: you say the build fails to load only if has a delay. But this code is never run (since it's the rolo code and you said it fails when the build you are loading has the delay)
16:26:12Mihailyes! And I have this mad case!
16:28:03pamaurycan you try it when you replace the sleep with something else. For example another lcd_putsf
16:33:16 Quit edhelas (Quit: Leaving.)
16:39:56MihailI comment line with sleep and uncomment with lcd_puts and have same problem. If I uncomment both lines - I still have same problem.
16:43:35 Join CrashBash-Kun [0] (~CrashBash@unaffiliated/crashbash-kun)
16:45:14 Join autofsckk [0] (~autofsckk@unaffiliated/autofsckk)
16:46:46autofsckkhi everybody, i have a little problem with my ipod 160G classic, i cant write or read the disk when i connect with usb, i tried to fix it with windows and it says it has to be formatted, and i also tried to fix it with linux but i couldnt :/ so what can i do? if theres a fix with linux i would prefer that because i dont have windows around jaja
16:49:09Mihailpamaury: If I add four lines "lcd_puts(0, 2, filename);" - it work again!
16:50:02Mihailautofsckk: Can you mount it in linux?
16:52:27pamauryMihail: ok, for it confirms that the problem is related to cache/alignment/dma, I had a very similar problem on imx
16:52:53pamauryOr buffer overflow
16:56:15MihailBut I still can't understand - why it work if I load it from internal storage?
16:59:04pamaurythat's a good point, I don't know :(
16:59:28 Join edhelas1 [0] (~edhelas@2001:1c02:1903:d800:1b:c303:7946:c98e)
17:02:03autofsckkMihail -> no i cant
17:03:01autofsckkMihail -> it works right to use it and hear music, but i cant read the disc from usb
17:03:25Mihailpamaury: Maybe rolo load it to memory bit different (corrupted)? It possible to add simple checksum to rolo or at early rockbox initialization?
17:04:35Mihailautofsckk: What you have in dmesg when player connected?
17:05:17pamaurywe already have a checksum
17:06:57pamaurythe only way the memory could be corrupted is something modifies the memory after the checksum, for example a dma engine still working, or some dirty cache
17:07:07pamauryat least I don't see anything else
17:08:03autofsckkit seems my disc is almost dead :( jajajaja <−−- Mihail
17:09:58Mihailautofsckk: You have same problem in OF?
17:10:40autofsckksorry whats of?
17:12:38Mihailoriginal firmware
17:13:09autofsckki dont know, i dont think so it was working all
17:13:24autofsckk but i have been using rockbox for a long time so i really dont remember
17:15:13Mihailyesterday we have report about mostly same problem:
17:21:10 Join xorly [0] (
17:25:45 Quit petur (Read error: Connection reset by peer)
17:26:47 Join PurlingNayuki [0] (
17:26:49Mihailpamaury: if I understand correct we don't check sum if CPU_ARM.
17:27:16pamauryare ou kidding me ?!
17:27:23*pamaury checks
17:27:38MihailLook again to rolo source ;)
17:28:49Mihailwe have to many #ifdef ...
17:30:20pamauryMihail: we do a checksum
17:30:37pamaurywe call LOAD_FIRMWARE that calls the correspond descrambling function
17:30:59pamaurythe standard one is load_firmware in firmware/common/rb-loader.c and it checks for checksum
17:31:53Mihailyou right
17:37:35autofsckkMihail -> any idea on how i can fix that error on my ipod?
17:39:59Mihailautofsckk: no, as I know about ipod mostly nothing.
17:41:25autofsckkah ok, thanks anyway Mihail :D
17:44:08 Join smoke_fumus [0] (~smoke_fum@
17:52:48***Saving seen data "./dancer.seen"
17:55:21Mihailpamaury: I move LOAD_FIRMWARE() before rolo_restart(). It load firmware but I still have same problem. Can you check is rolo_restart() and commit_discard_idcache() correct?
17:56:20pamauryMihail: what do you mean "before rolo_restart()" ? It is always before rolo_restart()
17:56:54MihailI mean one line above.
17:58:22Mihailso we do only rolo_restart() and commit_discard_idcache() after LOAD_FIRMWARE()
18:01:04pamauryI think the problem is elsewhere. I just realise now that rolo does not even ensure proper shutdown of rockbox
18:01:57pamauryit calls audio_stop() then storage_flush() then adc_close(), then ascodec_write_pmu(..) on amsv2, then disable irq
18:02:32pamauryso if you have any pending dma/timer/watchdog/lcd operation pending, who knows what can happen
18:05:56pamauryfor me the code should look like what we do in bootloaders: calling system_prepare_fw_start()
18:13:01Mihailbut we don't have system_prepare_fw_start() for AMS
18:17:36pamaurymaybe it is not needed, I don't know, I just find it a suspicious
18:17:41pamaury*a bit
18:25:39pamauryFor example, it seems entirely possible to me, with the current code that:
18:25:39pamaurya) main thread starts doing rolo_load()...
18:25:39pamauryb) but is preempted in some function (lots of functions indirectly call yield()) and another thread does a read/write operation and yield() on a wait
18:25:39DBUGEnqueued KICK pamaury
18:25:39pamauryc) rolo_load() finally disable IRQ, so it cannot be interrupted anymore and jump to the new image. But now there is a pending I/O operation
18:25:57pamauryI am taking the example of I/O but really it could be anything
18:26:06pamaurysome target use threads to handle inputs and buttons
18:28:11 Quit bluebrother^ (Ping timeout: 252 seconds)
18:28:40pamauryFor example, on imx233, there is a long standing bug that triggers when bootloader usb mode is used: around 1/3 of the time, the boot fails when loading the image
18:29:31 Quit krnlyng (Quit: krnlyng)
18:29:44Mihailok, I agree with your point. I should try disable all hardware which possible.
18:29:54 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
18:36:22pamauryyes, even on imx it's broken. We need something like system_prepare_fw_start() that properly shuts down the relevant hardware, it needs to be fixed on bootloaders and implement for rolo, or at least investigated
18:37:13 Quit JanC (Ping timeout: 276 seconds)
18:50:13 Join JanC [0] (~janc@lugwv/member/JanC)
18:52:17 Quit TheSeven (Ping timeout: 250 seconds)
18:52:19 Quit PurlingNayuki (Ping timeout: 246 seconds)
18:54:21 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
19:23:56 Join chrisb [0] (
19:28:26 Quit metaphysis (Quit: metaphysis)
19:49:16 Join Onj [0] (
19:49:41 Quit Mihail (Quit: - A hand crafted IRC client)
19:50:31 Quit Onj (Client Quit)
19:52:31 Quit CrashBash-Kun (Ping timeout: 246 seconds)
19:52:51***Saving seen data "./dancer.seen"
19:57:27 Quit ZincAlloy (Ping timeout: 268 seconds)
20:00:57 Join ZincAlloy [0] (
20:16:11 Quit pamaury (Remote host closed the connection)
20:19:02 Join krabador [0] (~eros@unaffiliated/krabador)
20:19:02 Quit krabador (Client Quit)
20:19:52 Join krabador [0] (~krabador@unaffiliated/krabador)
20:59:16 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
21:03:49 Join CrashBash-Kun [0] (~CrashBash@unaffiliated/crashbash-kun)
21:14:44 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
21:34:56 Join krnlyng [0] (
21:35:49 Join lebellium [0] (
21:46:51 Quit krnlyng (Quit: krnlyng)
21:49:11 Join krnlyng [0] (
21:52:53***Saving seen data "./dancer.seen"
22:11:24 Part chrisb ("rcirc on GNU Emacs")
22:21:01 Quit djukon (Ping timeout: 244 seconds)
22:36:42 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
22:40:39 Quit bertrik (Ping timeout: 250 seconds)
22:47:39 Quit CrashBash-Kun (Ping timeout: 268 seconds)
23:05:25 Quit pamaury (Read error: Connection reset by peer)
23:06:45 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
23:08:25 Quit pamaury (Read error: Connection reset by peer)
23:09:42 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
23:09:52 Join Mihail [0] (25d47d70@gateway/web/cgi-irc/
23:10:27 Quit pamaury (Read error: Connection reset by peer)
23:11:44 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
23:13:42Mihaillebellium: Did you try install version before frequency scaling to your clip zip?
23:14:52lebelliumMihail: nope, didn't have time (or didn't want it to get on my nerves again :D )
23:14:59lebelliummaybe this weekend
23:15:31 Nick suYin`OFF is now known as suYin (
23:17:05Mihailok, now we can boot from sd card if you update firmware on it (bootloader should be updated).
23:22:22lebelliumdoes that mean that if I get an ATA error, it will try to boot from SD card instead?
23:24:12MihailYes. But only if your player still live enough for one firmware update.
23:25:00lebelliumIt's still recognized by the computer so it shoudn't be an issue
23:26:39lebelliumI don't see any change in RButil in your commit
23:26:50lebelliumso I assume I can't use it to patch the firmware?
23:29:27MihailIt not for RButil directly but for bootloader which RButil should use.
23:30:13 Quit autofsckk (Remote host closed the connection)
23:30:34 Quit FSanches (Quit: Leaving.)
23:30:58lebelliumso I can use the current RBUtil 1.4.0
23:31:43 Quit bertrik_ (Remote host closed the connection)
23:32:06lebelliumI'll try it out in the next days and will give you a feedback
23:33:28lebelliumIf it works well and is totally transparent for the user, that means that every Sansa user should install this bootloader just in case... one day... :)
23:34:16MihailI not sure about RBUtil. Can clarify what we need to update bootloader trough RBUtil?
23:34:35Mihail*Can someone
23:35:45lebelliummaybe you can have a look at the previous bootloader releases here;a=summary
23:38:31 Quit xorly (Ping timeout: 240 seconds)
23:38:52MihailIt just source code. But I can't understand is we build bootloader automatically?
23:41:53lebelliumI just know that the user only provides the firmware file and Rbutil patches it automatically
23:42:00lebelliumbut I don't know the magic behind
23:42:10lebelliumhope someone around can explain it
23:50:34 Quit lebellium (Quit: ChatZilla 0.9.92 [Firefox 45.0.2/20160407164938])
23:50:34 Nick suYin is now known as suYin`OFF (
23:52:54***Saving seen data "./dancer.seen"
23:58:09 Join PurlingNayuki [0] (

Previous day | Next day