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 2013-11-17

00:00:46pamaury[7]: can you have a look at my driver, see if you spot anything ?
00:01:35pamauryI tried to use yours without modification but it didn't work, problem is I don't have any feedback, beside able to disconnect usb. Currently my driver gets the reset but the setup interrupt is never triggered
00:04:25 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
00:04:54wodzpamaury: enum_done is 'setup' irq?
00:07:29 Quit pamaury (Quit: No Ping reply in 180 seconds.)
00:08:15 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
00:09:37pamaurywodz: no
00:09:50pamaurysetup is in oepint
00:10:24pamauryand then you have a check a subflag in theory but I don't even get there anyway
00:14:54 Join pamaury_ [0] (~quassel@rockbox/developer/pamaury)
00:15:04 Quit y4n (Ping timeout: 272 seconds)
00:17:00 Quit wodz (Quit: Leaving)
00:17:03 Quit pamaury (Ping timeout: 245 seconds)
00:19:09 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
00:20:08 Quit pamaury_ (Ping timeout: 272 seconds)
00:29:21 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
00:45:08 Quit cmhobbs (Ping timeout: 248 seconds)
00:51:22 Quit nosa-j (Ping timeout: 252 seconds)
00:53:11 Quit ruskie (Ping timeout: 246 seconds)
00:55:57[7]pamaury: so you do get an enumdone irq but not an oepint one?
00:56:05 Quit Narod ()
00:58:02pamaury[7]: yes, iirc
00:58:26[7]ep0 usbactep bit set?
00:59:02pamauryit's set in enum done I think
00:59:05[7]or is that even hardwired?
00:59:29pamauryI think it's hardwired but I didn't take any chance ^^
00:59:39[7]are you just not getting an irq or is the setupdone bit never getting set?
00:59:46[7]could also be in irq mask problem
00:59:52[7]are you using dma?
01:00:31[7]could also be fifos or dma config, pio irq handling
01:01:04 Nick DormantBrain is now known as SuperBrainAk (~andy@
01:01:27pamauryit's supposed to be PIO
01:01:43pamauryI don't reset the core so I get the boot rom setup of the core which I know is correct
01:02:00pamaurythe setupdone bit is never set, I tried to busy loop on it
01:03:00[7]any pio copy irqs?
01:03:22[7]those would happen before setupdone
01:03:58pamauryI don't recall getting any no, but I'm working blind so it's tricky
01:04:30pamaurySo the expected events are: reset, enum done, transfer complete, setup done, ?
01:06:45 Quit ender` (Quit: I used to believe in reincarnation, but that was in a previous life.)
01:08:00 Join nosa-j [0] (~m00k@
01:14:15 Join pamaury_ [0] (~quassel@rockbox/developer/pamaury)
01:14:30 Quit pamaury (Ping timeout: 260 seconds)
01:16:24 Quit nosa-j (Ping timeout: 245 seconds)
01:16:59 Join shamus [0] (
01:17:08 Quit foolsh (Quit: foolsh)
01:20:25[7]pamaury_: no, you won't get a transfer complete for SETUP packets
01:21:09pamaury_ok that's what I thought, so I should get setup done directly right ?
01:21:13[7]you should get reset, enumdone, one or more nptxfempty and finally a setupdone
01:21:37pamaury_I haven't enabled nptxfempty at the moment
01:21:42[7]you need to handle the nptxfempty condition before you'll get the setupdone if you don't use DMA
01:21:53pamaury_why is that ?
01:21:56[7]the code basically offloads the DMA engine's work to the CPU using that IRQ for PIO mode
01:22:09[7]it basically issues "DMA requests" to your driver :)
01:22:42[7]so the nptxfempty IRQs take care of copying the data from the fifo to memory, so that it's available when setupdone is issued
01:23:06pamaury_ok, the setup is treated like a regular transfer
01:23:30[7]except that setupdone is asserted instead of xfercomp
01:23:37[7]and some other things with back to back setup packets etc.
01:24:14pamaury_ok, and if I don't pull the data off the fifo, it won't send setupdone most probably
01:24:20 Quit kugel (Read error: Operation timed out)
01:24:44[7]depending on the revision of the core there are two versions of how this PIO mode works
01:25:07pamaury_what are the differences ?
01:25:08[7]if you have the newer one, you can use the rxstsqlvl IRQ instead of the nptxfempty one
01:25:45[7]that's for the revisions that have dedicated TX fifos per endpoint
01:26:10pamaury_how can you tell which mode is used ?
01:26:32[7]er.... I'm talking nonsense
01:26:34[7]that was for TX
01:26:37[7]you're doing RX
01:26:39pamaury_I have the capability to hack some rknano program and change the VID:PID to any value, including a register
01:26:45pamaury_that's how I figured out it's cortex m3
01:27:09[7]you need to handle rxstsqlvl
01:27:12[7]if you do PIO
01:27:37[7]if it's M3 then I guess it's likely that it has the new version of the core
01:28:01[7]so enable rxstsqlvl IRQ, and handle it as follows:
01:28:02pamaury_so it's rxstsqlvl
01:28:53[7]read grxstsp, determine the endpoint from its chnum field, determine the number of words from the bcnt field, and copy that many words to whereever your transfer buffer for that input is
01:29:16[7]then deassert the IRQ as usual by setting that bit in gintsts
01:29:41[7]that's the RX part, which is the simple one
01:30:01[7]for TX it depends on the core version
01:30:58pamaury_great :-/
01:31:10[7]the old version uses nptxfempty, and you need to choose an endpoint to push data for yourself everytime that IRQ is asserted
01:31:40[7]for the new version, with dedicated TX fifos per endpoint, you can just use the emptyintr IRQ of those endpoints to push data
01:32:09 Join nosa-j [0] (~m00k@
01:33:08[7]the IRQs for TX needs to be masked to get rid of it when there's nothing left to send, and unmasked as soon as new data needs to be transmitted
01:34:08pamaury_ok, I'll try this tomorrow then
01:35:05[7]you need to read the data from DFIFO
01:35:12[7]which is OTG_BASE+0x1000
01:35:38***Saving seen data "./dancer.seen"
01:36:40[7]DMA is certainly much easier to get to work if the core supports it
01:37:49pamaury_from the code I saw in the stub, it seems to use PIO
01:43:44 Quit sakax (Quit: Leaving)
01:47:24 Quit nosa-j (Ping timeout: 252 seconds)
01:52:35 Join nosa-j [0] (~m00k@
01:57:48 Join kiwicam [0] (~quassel@
01:59:33[Saint]Hmmmmm...this is curious.
02:00:01pamaury_Yes ?
02:00:15[Saint]Every time, and I mean *every* time, I plug my Classic in to transfer some files it sets the dirty bit.
02:00:54[Saint]And its definitely unmounting cleanly.
02:03:28 Quit bertrik (Remote host closed the connection)
02:04:21[Saint]If I thin check with fsck.vfat, the dirty bit is the only thing there to clean up.
02:04:24[Saint]Its weird.
02:09:38[7]you mean linux sets it, or rockbox before/after entering USB mode?
02:11:52[Saint]I didn't think Rockbox was capable of this.
02:12:24[Saint]I assumed it was Linux's doing, as I didn;t think Rockbox had any way of knowing the state of the FS.
02:12:28[7]hm... no idea, but it seems to be somewhat trivial to implement
02:13:15[Saint]There's always been a firm "keep fsck-esque stuff out of the core" mentality.
02:13:38[7]if linux is doing it, fsck, mount, unmount, and fsck again should have it set during the second fsck
02:13:52[7]fsck != dirty bit
02:14:26[Saint][14:13:40] <[7]> if linux is doing it, fsck, mount, unmount, and fsck again should have it set during the second fsck <−− this, exactly this
02:14:44[7]rockbox provably ignores the bit, but should really be setting it to make a pc run fsck when necessary, e.g. after a menu+select reset while writing data
02:15:59[Saint]Its as though the mere act of inserting USb sets the dirty bit.
02:16:01[7]so unmount isn't clearing the bit... and you aren't by any chance doing unmount -l?
02:17:05[Saint]I switched out to doing it manually so I could rule out Nautilus doing anything weird. But it still happens.
02:17:07[7]so if you clear the bit, then mount and unmount, then fsck again, without disconnecting it in between, the second fsck says it's dirty?
02:18:29[7]hm... blame your kernel then
02:19:39[Saint]Its not /really. bothering me - its just one of those "THINGS THAT SHOULDN'T BE"
02:20:47 Quit ender| (Ping timeout: 256 seconds)
02:21:50 Quit amiconn (Remote host closed the connection)
02:21:50 Quit pixelma (Remote host closed the connection)
02:21:58 Join ruskie [0] (ruskie@sourcemage/mage/ruskie)
02:22:16 Join ender| [0] (krneki@2a01:260:4094:1:42:42:42:42)
02:23:02 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
02:23:02 Join amiconn [0] (amiconn@rockbox/developer/amiconn)
02:24:40 Nick cWiden is now known as tUasal (~random-na@
02:26:01 Nick tUasal is now known as freamh (~random-na@
02:54:42 Quit nosa-j (Quit: k)
02:55:29 Quit kiwicam (Remote host closed the connection)
02:56:33 Join nosa-j [0] (~m00k@
02:59:00 Nick freamh is now known as tuXasal (~random-na@
03:15:09 Quit pamaury_ (Ping timeout: 245 seconds)
03:35:39***Saving seen data "./dancer.seen"
03:47:30 Join kiwicam [0] (~quassel@
04:04:54 Quit kiwicam (Remote host closed the connection)
04:21:22 Quit mirak (Ping timeout: 245 seconds)
04:38:17 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:38:17 Quit pixelma (Disconnected by services)
04:38:17 Quit amiconn (Disconnected by services)
04:38:17 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
05:11:54 Quit Guest17648 (Quit: Hmmm...)
05:12:41 Quit [7] (Disconnected by services)
05:12:54 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:33:11 Quit saratoga (Quit: Page closed)
05:35:43***Saving seen data "./dancer.seen"
06:33:51 Quit [Saint] (Remote host closed the connection)
06:35:54 Join [Saint] [0] (~saint@rockbox/user/saint)
07:24:51 Join eternnoir [0] (uid13219@gateway/web/
07:28:48 Join redtape|coffee [0] (~random-na@
07:30:07 Join kiwicam [0] (~quassel@
07:32:06 Quit tuXasal (Ping timeout: 265 seconds)
07:35:46***Saving seen data "./dancer.seen"
07:39:43 Nick redtape|coffee is now known as Tuxi848 (~random-na@
07:50:09 Nick SuperBrainAk is now known as DormantBrain (~andy@
08:06:54 Quit Guest26236 (Ping timeout: 272 seconds)
08:10:11 Join Scall [0] (~chat@unaffiliated/scall)
08:30:54 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
08:33:21[Saint]Would anyone have anything against me removing that god-awful "nuke config.cfg if hold is on during boot" check?
08:34:08[Saint]Disabling it solely for the Classic seems uglier than disabling it all round.
08:34:56[Saint](The Classic needs it (IMO) as turning hold on, if the device is off, will boot the device - and boom - wiped config)
08:36:38[Saint]The number of times I have looked at my iPod and thought "Oop, better turn hold on" when I'm putting it away in my bag or whatever only to find I've nuked my settings...gah.
08:49:21 Quit eternnoir (Excess Flood)
08:49:39 Join eternnoir [0] (uid13219@gateway/web/
09:00:12 Quit Scall (Ping timeout: 272 seconds)
09:01:10 Join Scall [0] (~chat@unaffiliated/scall)
09:35:47***Saving seen data "./dancer.seen"
10:26:42 Join ender` [0] (
10:27:00 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
10:42:45 Join Narod [0] (
11:28:04 Join bertrik [0] (
11:28:13 Quit bertrik (Changing host)
11:28:13 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
11:32:24 Join dfkt [0] (OxO29A@unaffiliated/dfkt)
11:35:48***Saving seen data "./dancer.seen"
11:41:04 Quit mc2739 (Ping timeout: 272 seconds)
11:42:47 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
11:47:01 Join n1s [0] (
11:47:01 Quit n1s (Changing host)
11:47:01 Join n1s [0] (~n1s@rockbox/developer/n1s)
12:07:44 Join mortalis [0] (~mortalis@
12:18:38TheSeven[Saint]: turning hold on shouldn't wake it up
12:19:13TheSeventhat might be a sign of a dodgy worn out hold switch
12:23:23*copper doesn't want to try it
12:26:44*user890104 suggest chaging "nuke config.cfg if hold is on during boot" to "rename config.cfg to config.cfg.bak if hold is on during boot"
12:33:19[Saint]TheSeven: both the CEATA ones I have around me presently do it.
12:34:01 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
12:34:18[Saint]copper: if you're quick enough, you can turn hold off again as soon as emCORE boots but before ROckbox takes over.
12:40:44copperI'll just do this the proper way: backup config.cfg and then test
12:47:01 Quit DuperMan (Disconnected by services)
12:47:03 Join DuperManawy [0] (
12:51:09copper[Saint] TheSeven: turning hold on while my iPod Classic is off doesn't turn it on. Turning hold back off DOES turn on my Classic. The config file doesn't change or get deleted.
12:52:07[Saint]that's a "thin" one, yes?
12:52:23TheSeventhe power management circuitry should be the same
12:52:25copperyup, latest model
12:52:38 Join Guest17648 [0] (
12:52:54TheSevenI'm guessing that your hold switch is worn out and thus bounces a few times between off and on while switching
12:53:13[Saint]it must be a fairly common ailment then.
12:53:34[Saint]either that or I use devices in some weird way that consistently breaks it.
12:53:49TheSevenmy nano is also doing that once in a while, but not consistently (yet)
12:57:37[Saint]I think I remember accidentally nuking my config once or twice waking waking ipods up via the charger.
12:57:55[Saint](with hold on)
12:58:35[Saint]No, they can charge while sleeping and don;t wake up, thankfully.
12:59:23[Saint]Video, COlor, 4G, Nanos...
13:00:35TheSevenhm, I can reproduce your hold switch behavior on mine (waking up while turning hold on)
13:00:52TheSevenbut if I apply a bit of pressure to the switch while moving it, it doesn't happen
13:00:58TheSevenso it's definitely the switch in my case
13:01:14[Saint]so apparently the switches are really shitty - nice to know.
13:01:39[Saint]I shall stop using it, I guess.
13:03:11TheSevenhm... or check if we can somehow disable the hold switch wakeup so that you have to press a button as well to power it on
13:03:12coppernever happened to me
13:03:24copperI often turn hold on after I turn off my iPod
13:03:29TheSevenjust like we disabled powering up when plugging in USB
13:17:15 Quit DuperManawy (Ping timeout: 264 seconds)
13:19:22 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
13:20:51 Quit bertrik (Ping timeout: 264 seconds)
13:29:57 Join DuperMan [0] (
13:30:12copperthe hold switch on my Classic is quite stiff, btw
13:30:21copperwith good tactile feedback
13:34:18 Quit DuperMan (Ping timeout: 240 seconds)
13:35:32 Join DuperMan [0] (
13:35:49***Saving seen data "./dancer.seen"
13:36:17copperhey, you know what would be cool? A database activity indicator tag for themes.
13:41:03 Quit DuperMan (Disconnected by services)
13:41:05 Join DuperManawy [0] (
13:44:32 Join DuperMan [0] (
13:46:14 Quit DuperManawy (Ping timeout: 245 seconds)
13:49:22 Quit DuperMan (Ping timeout: 252 seconds)
14:05:23 Quit mortalis (Quit: KVIrc 4.3.1 Aria
14:24:06 Nick Tuxi848 is now known as Kality (~random-na@
14:26:28[Saint]*All* my Classics have this hold swith hardware bug thingy.
14:26:47[Saint]No wonder I thought it was normal behavior.
14:26:52copperall second hand?
14:27:13[Saint]Apparently the hold switch failing is a common thing.
14:27:16[Saint]Not all, no.
14:27:46[Saint]Two of them have only been owned by myself.
14:29:26[Saint]The new ones just feel like a crisper action, with the same behavior.
14:29:31 Join mortalis [0] (~mortalis@
14:29:37copperApple hardware is known to be occasionally faulty
14:30:25[Saint]Well, one is new new, and the other was a refurbished shop demo model so it never saw any rough treatment.
14:30:35[Saint]The other two are second hand and well used.
14:33:02[Saint]Maybe I'm just a pansy.
14:33:28[Saint]If I use the hold switch /quite/ forcefully, it works as intended occasionally.
14:35:24[Saint]I kinda don't want to play with them now. It would be a PITA to have it completely non-functional.
14:39:52coppermaybe they've silently switched components during production
14:42:01 Join lorenzo92 [0] (
14:42:29copperdidn't Samsung even switch SoCs in one of their galaxy phones once?
14:42:42copperwithout changing the model name / number
14:45:23copperI forget.
14:46:42[Saint]wildly offtopic - but NAND I think you're thinking of. Bricking themselves using trim iirc.
14:48:27copperGalaxy SII: they switched SoCs (different CPU and GPU) and still called it a Galaxy SII, but they did change the model number from i9100 to i9100G
14:48:44copperExynos 4210 to TI OMAP 4430
14:49:17copperI remember there were similar problems with Samsung Displays
14:49:34coppersame model number but different internal displays, with different performance
14:49:42copperbut yeah, OT :P
14:50:43lorenzo92pamaury: ping
14:51:41pamaurylorenzo92: pong
14:53:12pamauryTheSeven: ping
15:02:22TheSevenpamaury: pong
15:07:20pamauryI'm trying to make this usb thing work on rknano but no luck, is there something needed to enable rxflvl beside unmasking in GINTMSK ?
15:13:19 Quit Guest11127 (Ping timeout: 245 seconds)
15:21:46 Quit amiconn (Disconnected by services)
15:21:46 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
15:21:51 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
15:35:52***Saving seen data "./dancer.seen"
16:02:01 Join foolsh [0] (
16:04:32[Saint]copper: I'm interested in what made you say that.
16:04:57[Saint]Do you often get large bursts of seemingly inexplicable disk access too?
16:14:59 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
16:16:48 Quit [Saint] (Remote host closed the connection)
16:18:50 Join [Saint] [0] (~saint@rockbox/user/saint)
16:38:18TheSevenpamaury: I don't think so - this is the only thing that I spotted in my init code:
16:38:18TheSevenif (!data->use_dma) gintmsk.b.rxstsqlvl = 1;
16:38:52TheSeven(which is bit 4 in that reg)
16:47:17 Quit Guest17648 (Ping timeout: 246 seconds)
16:58:57copper[Saint]: sometimes my HDD activity icon lights up and I have no idea wth Rockbox is doing
17:00:07copperthat's when my Rockbox setup is semi-screwy though
17:04:09copperalso, is it normal that the database screen in the debug menu doesn't show any activity while the database is updating, rather than "initiating" itself?
17:06:49copperthough I guess it would show activity while it's actually adding / updating data, not when it's scanning the HDD for new / updated files
17:35:54***Saving seen data "./dancer.seen"
17:50:08 Quit nosa-j (Read error: Operation timed out)
17:54:34 Quit cmhobbs (Ping timeout: 245 seconds)
17:54:55 Join nosa-j [0] (~m00k@
17:55:44 Quit nosa-j (Excess Flood)
17:56:27 Join lebellium [0] (
17:56:27 Join nosa-j [0] (~m00k@
18:00:59 Quit Scall (Ping timeout: 268 seconds)
18:02:12 Quit nosa-j (Ping timeout: 245 seconds)
18:06:40 Join Scall [0] (~chat@unaffiliated/scall)
18:07:17 Join nosa-j [0] (~m00k@
18:11:49 Join Guest17648 [0] (
18:25:33 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
18:32:04 Quit cmhobbs (Ping timeout: 252 seconds)
18:34:45 Join amayer [0] (
19:09:13 Join Strife89 [0] (
19:14:25 Join lebellium_ [0] (
19:15:31 Quit lebellium (Ping timeout: 245 seconds)
19:15:37 Nick lebellium_ is now known as lebellium (
19:17:14 Quit lorenzo92 (Ping timeout: 264 seconds)
19:17:26 Nick bertrik_ is now known as bertrik (~quassel@rockbox/developer/bertrik)
19:25:36 Quit Guest17648 (Ping timeout: 252 seconds)
19:25:57 Join Guest17648 [0] (
19:35:58***Saving seen data "./dancer.seen"
19:40:18 Quit mortalis (Quit: KVIrc 4.3.1 Aria
19:50:39 Join lorenzo92 [0] (
19:53:42 Quit Guest17648 (Ping timeout: 272 seconds)
19:53:54 Join Guest17648 [0] (
19:58:30 Quit Guest17648 (Ping timeout: 272 seconds)
19:59:28 Join Guest17648 [0] (
20:01:49 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:26:28 Quit Strife89 (Ping timeout: 248 seconds)
20:33:38 Quit Guest17648 (Quit: Hmmm...)
20:36:45 Join Guest17648 [0] (
20:43:24 Quit y4n (Quit: Do you like hurting other people?)
20:48:30 Nick DormantBrain is now known as SuperBrainAk (~andy@
20:49:19 Join rela [0] (~x@pdpc/supporter/active/rela)
21:03:00 Quit foolsh (Ping timeout: 246 seconds)
21:04:28 Join Wardo [0] (
21:27:19fs-bluebotBuild Server message: New build round started. Revision 8b06039, 243 builds, 36 clients.
21:33:16fs-bluebotBuild Server message: Build round completed after 357 seconds.
21:36:02***Saving seen data "./dancer.seen"
21:40:18 Join Szczepancio [0] (5309e28d@gateway/web/freenode/ip.
21:40:38SzczepancioHi pamaury, are you there?
21:42:10SzczepancioI downloaded OF and renamed it to, and I putted it into e380 root directory. But after this, i still have an bootloader mode at usb connection.
21:42:46pamaurydid you reboot to OF ?
21:42:49lebelliumbecause you have to boot to OF to upgrade the fw
21:43:10lebelliumOF = play + back
21:44:16Szczepanciolebellium: At what state?
21:44:32SzczepancioTurned off, usb connection, turned on?
21:44:40pamaurylebellium: only back, not play
21:45:10pamauryideally you should have transfered in OF mode, but anyway you can still try to power OF without usb now
21:45:36Szczepanciopamaury: Easier english, ok?
21:46:02SzczepancioOF mode is official firmware, i booted it after placing at root directory an
21:46:03pamaurypower off rockbox, power on OF by holding back for a few seconds
21:46:27pamauryok, then power off, hold the back button and plug OF
21:46:40pamauryplug USB, sorry
21:47:16 Quit lorenzo92 (Remote host closed the connection)
21:47:16pamauryremove, eject
21:47:41pamauryand start over again: boot to OF in USB mode, copy, eject, reboot to OF without USB
21:48:02 Join lorenzo92 [0] (
21:48:19Szczepanciohmm, i think i did it without removing an
21:50:28Szczepanciomeh, i just put unconsciously usb cable off from my pc box
21:51:43Szczepanciomaybe the bootloader usb mode is normal? i dont know
21:52:03Szczepanciobut no, it's wanna load rockbox, so it's unnormal
21:53:35Szczepanciopamaury: are you sure that your method is 100% working?
21:54:51pamauryyes, are you following the instruction ? what is the problem ?
21:55:47K1773Rrockbox seems to detect album covers when there is a file named {C,c}over.jpg but not if the file is named album.jpg. can this somewhere be changed?
21:55:57 Quit rela (Quit: Leaving)
21:56:12AlexPK1773R: There is a list of allowable names in the manual
21:56:20Szczepanciothe e380 is still booting in bootloader usb mode
21:56:23Szczepancioor no
21:56:40Szczepancioit's booting at usb plug without holding any butttons in bootloader usb mode
21:57:01SzczepancioI wanted to restore fully official firmware for my dad, so u know.
21:57:13lebelliumbe sure you press "back" when you plug the USB cable
21:57:20AlexPK1773R: If you want to use any other names you'll need to edit the code and then build Rockbox
21:57:45K1773RAlexP: ACK, so i need to find out how to setup the build enviroment and build the iPodClassic src :)
21:58:08AlexPyes, if you don't want to use one of the existing options or embedded art (for some formats)
21:58:38gevaertsAnd then live forever with the stigma of being the one person who couldn't deal with the options that were already there :)
21:58:40Szczepancioahh ok i booted into of without usb bootloader mode
21:59:51K1773RAlexP: existing options -> is there an option to add album.jpg to be recognized without recompiling?
22:00:23K1773RAlexP: yea, thats what i thougd. so you ment existing options in terms of what currently is being recognized
22:00:38AlexPThe list from the manual
22:00:45AlexPThere are 8 different options or so
22:01:32 Quit bluebrother (Read error: Operation timed out)
22:02:28 Quit fs-bluebot (Ping timeout: 245 seconds)
22:03:36K1773Rthanks, in this case im just creating a bash script which does mv/cp album.jpg to cover.jpg
22:03:47 Join fs-bluebot [0] (
22:04:06 Join foolsh [0] (
22:05:24 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
22:13:08 Quit Scall (Ping timeout: 272 seconds)
22:13:41 Quit ruskie (Quit: ...)
22:15:00 Join ruskie [0] (ruskie@sourcemage/mage/ruskie)
22:16:20SzczepancioMeh, done.
22:16:38SzczepancioIt's having official firmware without bootloader usb mode. Thank's again, pamaury :)
22:16:39 Join Scall- [0] (
22:21:12 Join wodz [0] (
22:21:15 Join saratoga [0] (123e11e0@gateway/web/freenode/ip.
22:21:34saratogaalbum.jpg actually sounds pretty reasonable but i'm too lazy to add it to the source code and then fight with git to commit
22:26:27K1773Rsaratoga: its banshees default if you enable writing album covers to the folder were the album resides
22:33:38 Quit GodEater (Ping timeout: 252 seconds)
22:33:59 Quit Szczepancio (Ping timeout: 250 seconds)
22:36:12 Join GodEater [0] (~whoknows@2a01:348:6:2c6::2)
22:36:13 Quit GodEater (Changing host)
22:36:13 Join GodEater [0] (~whoknows@rockbox/staff/GodEater)
22:38:41 Quit Scall- (Quit: Bye bye)
22:43:57lonoxmontwould it make sense to try and order the other ipod cable for my car? since we know the current one uses iap over usb, do oyu think they would have used the same for both? i think on the website it said if you ahve an older ipod to use the other cable fromt he one i have...
22:44:12 Join Scall [0] (~chat@unaffiliated/scall)
22:44:19lonoxmontdunno if that means the ohter one uses serial or not
23:05:44 Quit Guest17648 (Quit: Hmmm...)
23:12:21 Join Guest17648 [0] (
23:16:37lebelliumpamaury: final result: 32.5hrs! I'm really impressed, that's even more than what Sony claimed (30hrs) and 3x more than Rockbox.
23:18:13 Quit amayer (Ping timeout: 246 seconds)
23:19:40 Quit krnlyng (Read error: Connection reset by peer)
23:21:29 Quit Guest17648 (Quit: Hmmm...)
23:21:34 Quit Wardo (Quit: Blarglarg)
23:21:45 Join krnlyng [0] (~liar@
23:24:18pamaurylebellium: i'm impressed too
23:26:59pamauryI'm still unsure of the reason, the only reason I'm thinking of is audio parameter
23:27:21pamauryIt seems the Fuze+ uses very different settings from the E370, from the register values I gathered last time
23:28:04pamaurylike biais adjustement and things I don't quite understand, maybe they have a great impact on battery life
23:33:00lebelliumbut you said you only got 21hrs with NWZ-E370 right?
23:35:06 Quit lorenzo92 (Ping timeout: 245 seconds)
23:35:15 Quit ender` (Quit: "HR?" "Human Resources." "In Brussels that kind of department is referred to as the Office for Personkind Enablement. Resources sounds like something you dig out of the ground." -- Peter F. Hamilton: Great North Road)
23:36:03***Saving seen data "./dancer.seen"
23:36:14 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
23:36:29pamaurylebellium: with OF yes
23:36:47pamaurywhich files did you use ?
23:36:51lebelliumAccording to Sony they should have the same battery life
23:36:57lebelliumMP3 VBR V0
23:37:04 Quit bertrik (Remote host closed the connection)
23:37:16lebelliumdon't they both have a 500mAh battery?
23:37:28pamaurybattery is not brand new
23:38:15lebelliummine neither
23:38:44pamauryanyway 21h is twice rockbox
23:39:03lebelliumbut I would assume they have the same battery capacity and that a 128x160 screen eats up less power than 240x320
23:39:14wodzpamaury: such huge battery drain could be some gpio driving some resistive load instead of being set in hiZ
23:40:13andypotterlonoxmont: Difficult to tell whether the other cable would be serial or not. Depending on how much a cable cost it could be worth a try.
23:41:33pamauryyeah but what ? The only gpio used at the moments are internal storage gate and backlight enable
23:41:55pamauryI tried to switch off backlight instead of just pwm'ing to 0 but it doesn't make a difference
23:42:13pamauryI tried to lowered the ADC for buttons, no difference either
23:42:51wodzpamaury: Are gpios as inputs by default?
23:43:23wodzyou could check in which state they are after sonys bootloader
23:43:26pamauryso except if the init stubs of the OF touch something (which I doubt), that leaves us with power and audio parameters only, or I'm missing something
23:44:25wodzWhat is the boot sequence? Is there something before your code?
23:44:31pamaurythe only major difference seems to be that the OF puts lcdif pins as input whenever the screen is off, and I simply turn the lcdif off
23:44:54pamauryonly OF init stubs, it inits sdram, it's not specific to the device really
23:46:16pamauryhum wait, I have an idea
23:47:11pamaurylebellium: can you tweak something and start a new rockbox bench for me please ?
23:47:40lebelliumyes, but I can only start the bench tomorrow since my battery is fully empty now :)
23:48:01pamauryah I see, mine should be charge faster than you since it's half full, i'll do it
23:48:44 Join amayer [0] (~amayer@
23:48:55wodzpamaury: What the idea?
23:49:01lonoxmontandypotter: i think ive seen them from 30 to 60 dollars
23:49:02pamaurystorage pins
23:49:14lonoxmontill see baout finding one at a reasonable price
23:49:29lonoxmontno way to know for sure itl be useful though
23:51:05lonoxmont scroll down to cables
23:51:20lonoxmont(yes i know its for the passat but they use the same harness)
23:52:38pamauryI need to figure out why hwstub is broken on Creative ZEN V too :'(
23:56:38 Quit kevku (Ping timeout: 260 seconds)
23:56:56 Quit kugel (Ping timeout: 252 seconds)

Previous day | Next day