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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2008-02-22

00:03:04toffe82aliask : take a look here
00:03:04 Quit tvelocity__ ("Αποχώρησε")
00:03:53 Join [1]Falafel [0] (
00:04:08toffe82aliask : you will find the definition for all the register for the mc13783 ( or almost all), I will try to browse all the file and update the wiki
00:05:09 Quit davina (Remote closed the connection)
00:05:34 Quit framo (
00:05:34 Quit freqmod_nx (
00:05:34 Quit ompaul (
00:05:34 Quit gevaerts (
00:05:34 Quit dan_a (
00:05:34 Quit csc` (
00:05:34 Quit Falafel (
00:05:34 Quit tchan (
00:05:34 Quit Nevtus (
00:05:34 Quit Neskaya (
00:05:34 Quit Galois (
00:05:34 Quit Addicted (
00:05:34 Quit n17ikh (
00:05:35 Quit martii (
00:05:38 Nick [1]Falafel is now known as Falafel (
00:06:27magmaniacJdGordon|w: someone already reported this at the flyspray metabug 8 months ago
00:06:35 Quit matsl_ (Remote closed the connection)
00:07:39 Join Neskaya [0] (
00:07:40***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
00:07:51 Quit Zarggg (Read error: 104 (Connection reset by peer))
00:07:59NJoinmartii [0] (
00:08:47NJoinfreqmod_nx [0] (
00:09:31 Quit lee-qid (Read error: 110 (Connection timed out))
00:09:49NJoingevaerts [0] (
00:10:00 Quit Rob222241 ()
00:11:43NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
00:11:47NJoinAddicted [0] (
00:12:59 Join Zarggg [0] (
00:15:15 Quit saratoga ("CGI:IRC")
00:16:24 Quit Neskaya (Killed by (Nick collision))
00:16:27NJoinframo [0] (
00:16:27NJoinNevtus [0] (n=Nevtus@unaffiliated/nevtus)
00:16:27NJoinompaul [0] (n=ompaul@gnewsense/friend/ompaul)
00:16:27NJoindan_a [0] (n=user@
00:16:27NJoincsc` [0] (n=csc@archlinux/user/csc)
00:16:27NJoinNeskaya [0] (i=featherd@wiktionary/neskaya)
00:16:27NJoinGalois [0] (
00:16:27NJoinn17ikh [0] (
00:16:29 Join Neskaya_ [0] (
00:16:35 Quit framo (Excess Flood)
00:16:36 Quit ompaul (Connection reset by peer)
00:16:39 Quit Neskaya (SendQ exceeded)
00:16:46 Join framo [0] (
00:16:54 Quit MajorC ()
00:17:08JdGordon|wmagmaniac: ok :( flyspray isnt really high on the priorities list though
00:17:50magmaniacJdGordon|w: no problem, it's just that it was my first flyspray report and I didn't want to get it wrong.
00:18:40 Join newnick [0] (n=63eef255@gateway/web/cgi-irc/
00:19:11newnick\nick bob
00:19:19 Quit newnick (Client Quit)
00:19:26 Quit n17ikh (
00:19:26 Quit Galois (
00:19:26 Quit dan_a (
00:19:26 Quit Nevtus (
00:19:26 Quit csc` (
00:19:46NJoinNevtus [0] (n=Nevtus@unaffiliated/nevtus)
00:19:47NJoincsc` [0] (n=csc@archlinux/user/csc)
00:20:01 Join rockbox_no0b [0] (n=63eef255@gateway/web/cgi-irc/
00:20:08 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
00:20:28NJoinn17ikh [0] (
00:23:57 Quit rockbox_no0b (Client Quit)
00:27:07 Join Rob2222 [0] (
00:31:10 Join cool_walking_ [0] (
00:31:12 Join jcollie [0] (
00:32:48 Join homielowe [0] (
00:34:02 Quit Nico_P (Remote closed the connection)
00:34:09 Quit midkay ("Leaving")
00:39:52 Join noob [0] (
00:40:28noobI have a question
00:40:32noobI consulted the manual
00:40:37noobabout this but I couldn't find it
00:41:04advcomp2019noob, we cant read minds
00:41:19noobI'm trying to look this up again
00:41:20noobI know
00:41:21noobhold on
00:41:38scorche|shand please use the enter less and put full thoughts on just one line instead of 3
00:41:47scorche|shs/enter/enter key
00:42:13noobit says "you must reinstall the original sana firmware before running sansapatcher"
00:42:28***Saving seen data "./dancer.seen"
00:42:51noobwhere do I find this original sansa firmware?
00:43:14noobI follded the link and it leads to a page then it directs me to the wiki
00:43:27BigBambiWhat DAP?
00:43:50scorche|shwhat device do you have?
00:43:54BigBambiAnd I assume you are using the manual for the install instructions?
00:43:57BigBambiSansa what?
00:44:05BigBambiversion 1 or 2?
00:44:29BigBambiAnd you are following the instructions in the manual?
00:44:37scorche|shdid you format your device?
00:44:40nooblet me explain what happened
00:44:59noobmy sansa stopped charging
00:45:03nooband I thought it was the battery
00:45:14nooband so I ended up buy one
00:45:16nooband it wasn't the case
00:45:24noobthe sansa caught a usb virus
00:45:57nooband my antivirus was able to pick it up and then deleted it and so now my battery won't recarge
00:46:04LloreanThe sansa can't catch a virus.
00:46:19noobI swear my anti virus picked up something on the sansa
00:46:28noobit was like an autorun usb virus
00:46:29noobareu sure?
00:46:34scorche|shvirii need to be specifically designed for whatever their intended target is
00:46:52noobwell this virus was for usb divices
00:46:56scorche|shthere could have been a virus for windows on there, but nothing could "infect" the sansa
00:47:03scorche|shit isnt that simple
00:47:19noobsomething is wrong with my sansa
00:47:20scorche|shevery device is different...not just "usb devices"
00:47:29noobit won't charge but it's not the battery
00:47:46noobok well, anyhow if that isn't the case then
00:48:04Lloreannoob: you can have a virus in a file on your sansa. but it'll be a computer virus, and won't affect the sansa itself.
00:48:07noobI was trying to uninstal the rockbox so I can reinstall everything
00:48:26noobok that's good to know
00:48:41 Quit noob ("-=SysReset 2.53=-")
00:48:43 Join noob [0] (
00:49:12noobso I tried to get rid of the bootloader but it's not happening
00:49:32noobI used that Rockbox_Installer_Full.exe
00:50:00noobbut it says something to this extent, "Mountpoint not found" please select from the list below
00:50:52noobI meant
00:51:02 Join alienbiker99 [0] (
00:51:11BigBambinoob: PLEASE use full lines
00:51:56 Quit Doomed_ (Read error: 104 (Connection reset by peer))
00:52:23 Quit JdGordon|w (" ajax IRC Client")
00:55:59noobok, I meant to say I used rbutilqt.exe and so I tired to uninstall my bootloader, but it keeps saying that sansa is not found, so I tried the drive button and I used "Autodect" But it says "mount point not found" and so I jsut manaully select the drive and push uninstall bootloader and same problem. So I tried to figure how to manually remove the boot loader and I found the Sansapatcher.exe. When I used the patch it says something l
00:58:41 Quit Nevtus ("Gone")
00:58:47scorche|shhow did you try to uninstall your bootloader?
00:59:06 Join JdGordon|w [0] (i=836b0049@gateway/web/ajax/
01:00:06noobusing rbutilqt.exe
01:00:13 Quit JdGordon|w (Client Quit)
01:00:36noobI think it jsut worked
01:00:39noobhold on
01:01:12 Join JdGordon|w [0] (i=836b0049@gateway/web/ajax/
01:01:45noobok it worked
01:01:54noobsorry about that
01:02:06 Quit ender` (" Microsoft is like a case of herpes. You can't kill it off entirely, just force it to go into a dormant stage for a while.")
01:02:44 Quit amiconn (Nick collision from services.)
01:02:51 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
01:03:16 Quit Arathis ("Bye, bye")
01:05:06 Join dan_a [0] (n=user@
01:05:12 Quit conando (Read error: 110 (Connection timed out))
01:07:17 Quit noob ("-=SysReset 2.53=-")
01:08:13*gevaerts thinks some people might be confused a bit
01:08:50 Quit spiorf (Remote closed the connection)
01:09:21JdGordon|wfairly likely
01:11:00 Quit mirak (Remote closed the connection)
01:11:59gevaertsAlthough that "usb virus" would explain a fair bit about our high-speed problem
01:15:07scorche|shRockAV? =P
01:17:12 Quit magmaniac ("Leaving")
01:18:23*gevaerts wonders why this usb controller sends a port status change interrupt, and then always claims there is a connection
01:19:23 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
01:23:42Evilnickgevaerts: congrats and a massive thankyou for your work on USB
01:24:36gevaertsEvilnick: thanks. There's still a lot to do to make it ready for default inclusion though
01:25:19Evilnickit's just nice to see the progress that you've made in such a short space of time
01:26:16 Quit ompaul (Client Quit)
01:27:50 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
01:28:07 Join Travis__ [0] (
01:28:19 Nick Travis__ is now known as Sivart0 (
01:28:39Sivart0hello everybody :)
01:37:02*gevaerts decides that this code rework won't be done today, and goes to sleep
01:37:07 Quit gevaerts ("ZZzz..")
01:42:58 Quit Febs (Read error: 110 (Connection timed out))
01:44:59 Quit Sivart0 ("ChatZilla 0.9.81 [Firefox]")
01:45:19 Quit Falafel (Read error: 104 (Connection reset by peer))
01:54:15 Part toffe82
02:04:33 Nick Neskaya_ is now known as Iaborant (
02:05:24 Nick Iaborant is now known as Neskaya (i=featherd@wiktionary/neskaya)
02:06:34 Quit dan_a (Read error: 110 (Connection timed out))
02:12:56 Join asn [0] (
02:13:30asnHow can I *format* my ipod? I'm getting many I/O errors, and it fails on most rsyncs. I think that the filesystems must be corrupted, since I'm using it for over 3 years.
02:16:38asnI can't just wipe the whole thing and start anew?
02:17:25scorche|shyou can, but why wouldnt you want to just check it first and see if you can just avoid all that?
02:18:10asna) I don't care about the stuff currently in. b) Just for the heck of it.
02:18:44scorche|shsearch for "manual restore" in the wiki...i think it is IpodManualRestore, but i am not sure
02:19:00asnThank you.
02:23:04 Part pixelma
02:31:34asnI also don't know if I have the 20* or the 13* version of iPod 5G
02:35:58 Join midkay [0] (n=midkay@rockbox/developer/midkay)
02:41:51 Quit hannesd (Read error: 113 (No route to host))
02:42:32***Saving seen data "./dancer.seen"
02:54:20 Quit cars ("CGI:IRC (Ping timeout)")
02:57:42 Join fasmaie [0] (
03:03:59 Join Cardini [0] (n=189fceb0@gateway/web/cgi-irc/
03:04:56CardiniExcuse me, but what I read on Tracker suggests that all RTC players now have an alarm feature. On my Sansa, however, I can find none.
03:06:11JdGordon|wthe tracker either lied or you were looking at a patch
03:06:16JdGordon|wor a request
03:07:36CardiniThank you. So...I've read that one of the patches, if used alone, works on Sansa. Can you refer me to a resource on compiling my own OS + Patch?
03:08:12JdGordon|wsee if this works...
03:08:20JdGordon|wlogbot: explain CrossCompiler
03:08:35JdGordon|wexplain CrossCompiler
03:08:45JdGordon|warg... look at the wiki
03:08:57CardiniI shall. Thank you.
03:09:19 Quit asn (Read error: 104 (Connection reset by peer))
03:11:23 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
03:17:02 Join toffe82 [0] (
03:17:44 Quit BitTorment__ ("Do directly to guantanamo bay. Do not pass go. Do not collect your human rights.")
03:20:06 Quit fasmaie (Read error: 113 (No route to host))
03:20:36 Quit JdGordon|w (" ajax IRC Client")
03:25:46 Quit Isolinear (Read error: 110 (Connection timed out))
03:27:17 Quit toffe82 (Read error: 104 (Connection reset by peer))
03:28:52 Join toffe82 [0] (
03:35:02 Quit DerDome (Nick collision from services.)
03:35:03 Join DerDom1 [0] (
03:35:13 Nick DerDom1 is now known as DerDome (
03:35:33 Quit Rob2222 (Read error: 110 (Connection timed out))
03:35:34 Join Rob222241 [0] (
03:36:01 Quit toffe82 (Read error: 104 (Connection reset by peer))
03:37:18 Join toffe82 [0] (
03:47:35 Join Rob2222 [0] (
03:51:32 Quit toffe82 (Read error: 104 (Connection reset by peer))
03:54:13 Quit Rob222241 (Read error: 110 (Connection timed out))
03:54:34 Join toffe82 [0] (
03:59:13 Join DaCapn [0] (
04:03:38 Quit Cardini ("CGI:IRC")
04:04:25 Join z35 [0] (
04:09:29 Quit csc` (Connection timed out)
04:12:22 Quit DaCapn (Read error: 104 (Connection reset by peer))
04:25:15 Quit BlakeJohnson86 ("Leaving.")
04:27:57 Quit Thundercloud (Remote closed the connection)
04:30:50 Nick fxb is now known as fxb__ (
04:32:44 Join Davide-NYC [0] (
04:32:46 Join BlakeJohnson86 [0] (
04:34:59 Join Doomed_ [0] (
04:36:20 Quit alienbiker99 (Read error: 104 (Connection reset by peer))
04:39:49 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:40:04 Join miepchen^schlaf [0] (
04:42:35***Saving seen data "./dancer.seen"
04:43:29 Quit DerDome ("Leaving.")
04:44:32 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
04:47:55 Quit Doomed_ (Read error: 104 (Connection reset by peer))
04:54:51 Join BlakeJohnson87 [0] (
04:57:44 Join alienbiker99 [0] (
04:59:33 Quit Davide-NYC ("ChatZilla 0.9.81 [Firefox]")
05:01:12 Quit nicktastic (Read error: 110 (Connection timed out))
05:12:17 Quit BlakeJohnson86 (Read error: 110 (Connection timed out))
05:12:24 Join Galois [0] (
05:19:26 Join B00daW [0] (
05:26:48 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
05:27:31 Join Llorea1 [0] (n=DarkkOne@
05:29:52 Join Makuseru [0] (
05:30:08Makuseruhow can a clear the Database on a Toshiba Gigibeat?
05:32:40krazykitdelete the *.tcd files in the .rockbox directory
05:36:46 Quit Horscht ("Snak 5.3.3 Unregistered copy. Evaluation period is over. Program will now quit. Thanks for using Snak.")
05:39:27Makuseruthank you
05:44:43 Quit Llorean (Read error: 110 (Connection timed out))
05:49:30 Quit nplus (Remote closed the connection)
05:50:41 Quit Makuseru (Remote closed the connection)
05:55:12 Quit toffe82 (Read error: 104 (Connection reset by peer))
05:55:12 Join toffe82_ [0] (
05:55:21 Nick toffe82_ is now known as toffe82 (
05:56:28 Join toffe82_ [0] (
06:09:48 Join DC1 [0] (
06:11:01 Join cars [0] (n=451de957@gateway/web/cgi-irc/
06:11:33carsWhen I shut Rockbox down, my 2nd gen Ipod immediately reboots and I have to restart it in using the Apple OS. Is this normal or is there some way I can workaround this behavior?
06:12:22 Join JdGordon_ [0] (
06:12:48 Quit JdGordon (Read error: 110 (Connection timed out))
06:12:50 Quit Llorea1 ("Leaving.")
06:14:26 Quit toffe82 (Read error: 110 (Connection timed out))
06:22:38 Join toffe82__ [0] (
06:22:47 Nick toffe82__ is now known as toffe82 (
06:36:21 Quit DC1 ("If Obi-wan ain't home then I don't know what the fsck we're gonna do. I ain't got no other connections on Tattooine.")
06:42:34 Quit toffe82_ (Read error: 110 (Connection timed out))
06:42:39***Saving seen data "./dancer.seen"
06:49:09midgeycars: thats normal. the 2nd gen can't shut down and rockbox lacks a sleep mode. i believe amiconn might be working on an implementation
07:12:44 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
07:20:06 Join flyback [0] (
07:20:17flybackany progress on a sansa m series port
07:20:27flybackI have fallen in love with my dad's 512 I got him and borrowed
07:20:42flybackthe 1 gig are on sale $15 recertified
07:20:54midgeylooks like you've just volunteered!
07:21:18flybackI would help with a port for this $7 player I go
07:21:22flybackjwin :P
07:22:22midgeyyou might be able to convince some devs to go out in buy the player, but most porting work is done by people who own the player and want to see rockbox on it
07:22:57flybackyeah if I get my own I might do it
07:23:02flybackbut I won't brick my dad's
07:23:03midgeyyou could ask around on some sansa forums to see if you can round up some people will to try and start a port
07:23:04flybackalready did that once
07:23:13flybackwell I saw some stuff on the web page
07:23:18flybackduimps of the chip etc
07:23:22flybackjust curious
07:23:25flybackI am not asking for anything
07:23:48flybackeven the stock firmware isn't bad
07:23:57flybacknothing like a juicebox
07:24:01flybackor jwin
07:24:02midgeyas far as i know, no one has taken much of a look at it other than identifying some chips inside
07:24:12flybackwith a random function that always picks the same songs
07:24:18flybackthe sansa is 100x better
07:24:50midgeyits running on a telechips chip so it might share some things in common with the other upcoming telechips ports
07:25:00flybackthere we go
07:25:10midgeythats probably where the similarities end
07:25:15flybackthink I will get myself one and one for a late bday present for my sister
07:25:22flybackeven stock they aren't bad for $15
07:25:27midgeybut anyway i'm off to bed
07:27:20 Quit homielowe (Read error: 104 (Connection reset by peer))
07:28:48 Join homielowe [0] (
07:31:10 Nick JdGordon_ is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
07:31:51 Join toffe82___ [0] (
07:35:01flyback <−− shame this series has a wtf ending
07:37:10advcomp2019flyback, that is off the topic
07:37:56 Join conando [0] (
07:39:02flybacksorry I did a global msh
07:39:44 Quit toffe82 (Read error: 110 (Connection timed out))
07:41:21 Quit perrikwp (" ajax IRC Client")
07:45:56 Join DaCapn [0] (
07:54:38 Quit cars ("CGI:IRC (Ping timeout)")
08:01:50 Join GodEater [50] (n=bryan@rockbox/staff/GodEater)
08:07:46 Join My_Sic [0] (n=3ea13b0f@gateway/web/cgi-irc/
08:09:12 Part toffe82___
08:15:47 Quit cool_walking_ (" ")
08:27:27 Quit My_Sic ("CGI:IRC")
08:27:33 Join My_Sic [0] (n=3ea13b0f@gateway/web/cgi-irc/
08:29:13 Quit BigBambi (Remote closed the connection)
08:37:07 Join Rob222241 [0] (
08:38:32 Join spiorf [0] (
08:42:40***Saving seen data "./dancer.seen"
08:48:21 Nick B00daW is now known as B--daW (
08:54:43 Quit fyrestorm (Read error: 104 (Connection reset by peer))
08:55:20 Quit Rob2222 (Read error: 110 (Connection timed out))
08:59:41 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
09:04:32 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:04:39 Join petur [50] (n=petur@rockbox/developer/petur)
09:05:38 Quit spiorf (Remote closed the connection)
09:11:20 Join CaptainSquid [0] (
09:21:50 Quit DaCapn (Read error: 104 (Connection reset by peer))
09:22:38 Join DaCapn [0] (
09:29:19 Join anubisice [0] (
09:31:33anubisicegood evening, I was wondering, after looking at the WPS page, are there WPS for a Sansa C series?
09:31:49GodEaterthere are - but not published anywhere useful
09:33:16anubisiceok, thanks −− i see one in the forum, ill rummage there
09:37:29 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
09:38:49 Quit My_Sic ("CGI:IRC")
09:41:29 Join OlivierBorowski [0] (
09:45:35 Quit CaptainSquid (Remote closed the connection)
09:47:07 Quit anubisice (" HydraIRC -> <- Go on, try it!")
09:49:58 Quit jhulst (Read error: 110 (Connection timed out))
09:50:46 Quit jhulst_ (Remote closed the connection)
09:54:08 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
09:57:45 Join BenniBoya [0] (n=chatzill@
09:58:27BenniBoyahey, what do i have 2 change to stop rockbox clearing when it boots with the hold switch on
09:58:33BenniBoyafor ipod nano
09:58:52scorchethe code
09:59:06BenniBoyaany idea what file?
09:59:14BenniBoyaor shouldni just surf
10:00:21 Quit jhulst_ (Remote closed the connection)
10:00:58scorchei would likely have to do the same processes that you could to find it precisely, so if i were you, i would either make handy use of grep or just wait around until someone comes around who happens to know offhand
10:01:34BenniBoyawhats that?
10:05:19BenniBoyakk thx
10:05:26BenniBoyaim in ubnut and cant dind my path file
10:05:39GodEateryour path file ?
10:05:41GodEaterwtf is that ?
10:05:55BenniBoyauuh i ran
10:06:08BenniBoyaat the end it said add the arm-elf dir to you "path file"
10:06:16BenniBoyai thinks its ment 2 be in etc but it isnt
10:06:37 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
10:07:02GodEaterhmm - actually - that's completely the wrong file
10:07:42 Join hannesd [0] (
10:08:08BenniBoyait said path file
10:08:15BenniBoyai had to o the same in my old distro\
10:08:42 Join ChristineTham [0] (n=7cbd2209@gateway/web/cgi-irc/
10:08:49markunBenniBoya: I've added the path in my ~/.bashrc file
10:09:07markunhi ChristineTham
10:09:14ChristineThamdoes anyone know how to open a gigabeat x60
10:09:44ChristineThami have unscrewed the three screws that i can find, but the back casing still won't budge
10:09:50markunI think toffe might know, but he's not here right now
10:10:24BenniBoyayo markun, wheres that?
10:10:38GodEaterBenniBoya: ~/ means your home directory
10:10:48GodEaterBenniBoya: the file you wanted is actually apps/main.c too
10:12:57 Quit BenniBoya ("ChatZilla 0.9.81 [Firefox]")
10:14:38ChristineThami'm looking at the pictures of a disassembled X60 on the gigabeatinfo wiki page, and apparently there are a few hinges on the back cover
10:17:26 Quit jhulst_ ("Konversation terminated!")
10:22:24 Join fyrestorm [0] (
10:24:14markunGodEater: I thought he was asksing where to add the path to the cross compiler binaries..
10:26:04GodEaterhe was, and he was also asking where to change the hold switch behaviour for resetting your settings
10:27:22 Join FOAD_ [0] (
10:27:23 Join tvelocity [0] (n=tony@
10:30:11 Join hamdiya [0] (i=c3e5ecd8@gateway/web/ajax/
10:30:42 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
10:31:51hamdiyais there a tag separator for wps files?
10:35:24GodEateryou asked that yesterday, and the answer is still no
10:36:04hamdiyanever got an answer afair
10:37:07pixelmaI don't even understand what a tag separator should be...
10:39:13hamdiyalike instead of having several %xl (e.g.) on several lines
10:39:29hamdiyawhy not be able to put them all in one line but seperated by a special character
10:41:10pixelmaok, understood now - but GodEater is right, there is none
10:42:29pixelmaand I also don't see a benefit of such a thing, putting them on different lines make it more readable for me
10:42:41***Saving seen data "./dancer.seen"
10:44:46 Quit FOAD (Read error: 110 (Connection timed out))
10:44:47 Nick FOAD_ is now known as FOAD (
10:50:03 Join Obie [0] (
10:51:21 Part Obie
10:57:30 Quit ChristineTham ("CGI:IRC (EOF)")
11:16:08 Join Axio [0] (
11:21:45 Quit hamdiya (" ajax IRC Client")
11:28:49 Join barrywardell [0] (
11:29:52 Quit scorche (Read error: 104 (Connection reset by peer))
11:37:38 Join BitTorment [0] (
11:40:26 Quit DaCapn (Remote closed the connection)
11:42:30GodEater <−− USB throughput measurements from a 5.5G 80GB ipod - if anyone is interested ;)
11:43:33 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
11:45:26preglowsomething makes me think they're using full speed
11:45:51LinusNi guess you mean high speed
11:46:14GodEaterinteresting difference though between EDM and the full OF
11:46:38*GodEater would be interested to see the same test from other PP devices
11:47:31GodEaterI might go start a wiki page for other results
11:56:08 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
11:56:39preglowLinusN: a just gathered from the name that you didn't get any higher than full-speed...
11:56:53preglownice to hear there's another badly named thing out there
11:57:02LinusNit's really silly
11:57:04 Join Febs [0] (
11:58:28LinusNdamn, the ipod fm radio remote is totally dead when i connect it
11:58:47LinusNand i still can't transmit on the serial port :-(
11:59:37 Join MethoS- [0] (
12:00:15GodEaterfeel free to add to it
12:00:44preglowi'd just be depressed if i did
12:00:57preglowlike anything relating to ipod nano ata performance does
12:01:47GodEaterI thought that just sucked in EDM ?
12:04:57rasherpreglow: USB 3.0 will introduce super speed, which will be higher than both full and high speed!
12:05:35rasher(not joking)
12:07:16fyrestormwhats after super?
12:10:52 Join Thundercloud [0] (
12:13:32rasherI suppose USB 4.0 will then introduce two speeds "regular" and "extra", both of which will be faster than super
12:13:51rasherJust to make sure sanity stays at a minimum
12:14:11scorchefull, high, super, regular, extra....
12:15:07rasher(that was a joke though)
12:15:34scorcheof course, but that doesnt mean it is unlikely due to current trends..
12:19:43GodEaterrasher: you forget "ribbed for her pleasure" :)
12:34:09 Join BenniBoya [0] (n=chatzill@
12:34:17 Join desowin [0] (n=desowin@atheme/developer/desowin)
12:34:28BenniBoyayo, can someone help em with patching, i dont flly understand it
12:34:28preglowfull, high, super, regular, extra, ribber for her pleasure
12:34:35preglowyou know, that sounds plausible, and that's the scary part
12:34:42preglowribbed, even
12:35:01GodEaterthank god *someone* thinks so to
12:36:10 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
12:36:41BenniBoya`can someone explain to me how t patch?
12:36:56GodEaterwe heard you the first time
12:38:28 Quit linuxstb (Nick collision from services.)
12:38:37 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
12:42:20 Join houbysoft [0] (
12:42:36BenniBoyacould someone just explain the paramaters of patch for me even?
12:42:43***Saving seen data "./dancer.seen"
12:42:59preglowpatch -p0 < patchname
12:43:09preglowthere are sources on the web for this kind of info, like google
12:43:32linuxstbAnd here -
12:44:11BenniBoyamust i be in the directory of what i am patching?
12:44:26 Quit XavierGr (Nick collision from services.)
12:44:37 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
12:44:49linuxstbOpen up the patch file in a text editor - you'll see the filenames that the patch is modifying. So your location needs to match those filenames.
12:45:07linuxstbe.g. if the patch contains "apps/main.c", then you'll need to be the same directory that "apps" is.
12:45:49BenniBoyakk gotya
12:53:03BenniBoyacan some1 remind me wat the path command is
12:53:09BenniBoyaw8 dw i got it
12:53:45GodEatercan you *PLEASE* speak english
12:54:25BenniBoyacould you remind me of what the command your ent to put into .bashrc to get arm-elf to work?
12:55:16GodEaterthat would depend entirely on where you installed your arm-elf toolchain to
12:55:47BenniBoyai used
12:55:54BenniBoyarunning from my build folder
12:56:50 Quit Siku () offers two choices for install location
12:58:33 Join webguest88 [0] (n=d92dc1e6@gateway/web/cgi-irc/
12:58:33BenniBoyait did ti automaticly for me
12:58:49BenniBoyai just asked me to add the directory to path
12:59:54GodEaterwell you can try "export PATH=$PATH:/usr/local/arm-elf/bin"
13:00:37 Quit webguest88 (Client Quit)
13:00:39BenniBoyathats the one neede
13:00:49BenniBoyaso type export + the path in terminal?
13:01:04linuxstbType everything GodEater put between quotes.
13:01:08BenniBoyakk ty
13:01:17BenniBoyafrom ym build folder or anywhere
13:01:26linuxstbFrom anywhere.
13:01:33BenniBoyait be perminent?
13:01:48linuxstbBut it will be forgotten next time you log in, so add the line to ~/.bashrc to make it run each time you log in.
13:02:43BenniBoyai just added it to the bottom, will that do
13:03:21 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
13:03:54BenniBoyakay thx
13:03:57BenniBoyaokay thanks
13:04:01BenniBoyai fogot the english thing
13:06:14 Quit BenniBoya ("ChatZilla 0.9.81 [Firefox]")
13:07:28 Part houbysoft
13:08:19 Join dan_a [0] (n=user@
13:12:07 Join j3_hsr [0] (
13:12:36 Join BigBambi [0] (n=86ceaf37@rockbox/staff/BigBambi)
13:12:45j3_hsrare their any plans on developing a port for the 6th gen ipod classic?
13:12:52GodEaterj3_hsr: no
13:12:58BigBambiFebs: you there?
13:13:06j3_hsrGodEater: ok thanks
13:13:13j3_hsrany specific reason?
13:13:20GodEaternone of us own one
13:13:40BigBambiFebs: Cqn you change people's names of the forum?
13:13:40 Join Jonny88 [0] (
13:13:47 Part Jonny88 ("Kopete 0.12.6 :")
13:13:49pondlifeBigBambi: Maybe it's his real name. poor guy
13:14:04BigBambipondlife: Poor guy indeed
13:14:23FebsI don't think so, but let me check and see. I assume if I look at the forum that the person will be obvious?
13:14:41j3_hsrok thats a reason :)
13:14:46BigBambiNot appropriate for the forum IMO (certainly in the UK anyway)
13:15:12 Quit j3_hsr (Client Quit)
13:15:45*linuxstb didn't even notice the name when replying...
13:15:59FebsI can't change the username. Llorean or one of the other forum admins would have to do it.
13:16:07pondlifeBigBambi: He's probably from Newark, no?
13:16:19BigBambiLinusN fills that criteria I believe
13:16:26BigBambiLinusN: ping!
13:16:28*pondlife can't think of one without the other... sorry
13:16:40BigBambipondlife: Me non understqnd
13:16:48*pondlife like anagrams...
13:16:52*BigBambi apologises for that sentance
13:16:58BigBambipondlife: aha
13:17:11LinusNBigBambi: pong
13:17:26pondlifeNow you have the gift, you will never be able to visit Newark again.
13:17:34BigBambiLinusN: Could you change this person's nalme on the forums please?;u=16378
13:17:47BigBambiIt isn't appropriate IMO :)
13:17:59LinusNis it such a big deal?
13:18:08BigBambiIt is pretty rude in the UK, yes
13:18:19LinusNto whom?
13:18:31BigBambiA lot of people, kids etc.
13:18:33pondlifeTo the poster himself, I'd say
13:18:46BigBambiIt is swearing, and I don't think we should promote that
13:19:01GodEaterit'd be akin to someone calling themself "fucker"
13:19:04LinusNi'm not sure letting him have his silly name is "promoting"
13:19:13BigBambiIt isn't just silly
13:19:40BigBambiAs GodEater says, it is akin to twat or fucker or ...
13:19:44linuxstbGodEater: Regarding USB speed testing, maybe amiconn's "test_disk" plugin could be compiled as a standalone app and run on the host OS?
13:19:54pondlifeI don't find it particularly offensive, but then I don't find swearing offensive. Only hate-speak.
13:20:01GodEaterlinuxstb: probably yes
13:20:08GodEaterI'll try it with just a plain dd in the meantime though
13:20:16BigBambipondlife: Me neither, but many many people do
13:20:24pondlifeThat's their problem ;)
13:20:28BigBambiIt isn't allowed on daytime TV for instance
13:20:39LinusNwell well, i changed it now
13:20:46BigBambipondlife: Sure, but I don't think it gives a good impression on the forums
13:21:13BigBambiLinusN: Cheers - I suspect it is worse in the UK than the US, and just think it gives a bad impression
13:21:25pondlifeLinusN: You could have gone for wonker or winker... ;)
13:21:32pondlifewunker, even
13:21:35BigBambiheh, I love the new one :)
13:21:36linuxstbLinusN: BTW, my build server is alive again, so you can re-enable it if that hasn't been done already (it made a very large video recording in the wrong partition whilst I was away and filled up the disk...).
13:21:54LinusNi honestly think our constant arguments regarding guidelines gives a lot worse impression
13:22:11LinusNwow - that wasn't very good english
13:22:15linuxstbLinusN: I agree
13:22:28BigBambiI don't disagree, but that is no reason to leave it there
13:22:35linuxstbBut on the other hand, we need some policing to avoid the forums becoming anarchic.
13:23:02LinusNlinuxstb: activated again
13:23:16pondlifeCan any H300 users here repro the bookmark or resume failures?
13:23:19BigBambiI tend to agree on the rules incidently
13:23:34LinusNi had some awful bookmarking issues on my ipod the other day
13:23:59LinusNthe resume/bookmark code is totally wacko
13:24:05pondlifeI'm more concerned with the resume code, as bookmarking is probably a layer on top of that
13:24:11pondlifeIt is
13:24:22LinusNis it repeatable in the sim?
13:24:34pondlifeI can't make it misbehave at all here.
13:24:37pondlifeSim or target
13:25:03LinusNsome of my bookmarks never show up in any list
13:25:19LinusNand it sometimes resumes from the beginning
13:25:34LinusNthose two issues are probably unrelated though
13:25:35pondlifeBookmarking has long been buggy - but resume seems to be a new-ish (MoB?) issue.
13:25:54LinusNsome claim it has been broken since mob
13:25:55pondlifeIf playback is incapable of resuming correctly, it would affect both
13:26:05pondlifeIndeed, it probably has
13:26:19pixelmaFebs: nice thingy ;)
13:26:21pondlifeThere is some factor I'm obviously missing
13:26:53linuxstbSomething that has struck me when I've been implementing codecs is that the resume info seems to be codec-specific. IIRC, for MP3 it stores the offset in bytes (presumably to deal with the MP3 VBR issues), and for most other codecs, seems to store in ms. I'm not sure this is always correct or consistent. i.e. some kind of little "resume API" with codec-specific resume info might be helpful.
13:28:17pondlifeThere's a problem where the offset exposed from playback.c to the outside world is the WPS offset (i.e. the position of the position slider). This is stored for resume, but on resume audio_play takes the value as a codec offset. Not quite the same thing
13:28:21pixelmaLinusN, pondlife: resume is broken for me in *some* folders on my c200 (which then exhibit the "skip 3rd track feature") since the spinlock commit (r16105 IIRC) and I couldn't figure out what determines such a broken folder...
13:28:55pondlifeHmm, I have a fairly flat and simple folder structure. <artist>/<album>/<track>
13:29:17pondlifeI wonder if I'm lacking variety, or depth. No smart comments, please.
13:29:29*pondlife hopes GodEater's not around
13:29:39*BigBambi deletes the half written line...
13:30:44pixelmaI only found out that the more RAM Rockbox uses the more folders are broken - I am now running a logf build and even more folders are broken. I put some infos in FS #8520
13:31:07pondlifeHmm. What sort of buffer size do you have?
13:31:22*pondlife wonders if it's a good old buffer wrap issue...
13:31:39pondlifeI'm down to a shocking 25MB here
13:31:51pondlifeWithout voice
13:32:49pixelmawhile checking for that I found out that voice file present or not does not have an influence on what is shown as buffer size in the rockbox info...
13:33:43pixelmathe current buffer size is reported as 29,741MB in the info screen
13:34:16GodEaterthat's a big buffer
13:34:30preglowperhaps i should change my name to something demeaning as well
13:34:52pondlifepixelma: IIRC the voice buffer is only loaded if you have voice menus or .talk clips enabled
13:35:43LinusNpreglow: how about preblow :-)
13:35:49pondlifeI wish for a debug menu option to list memory usage - aka "where did my buffer go?".
13:35:57preglowLinusN: you know, i've gotten that a lot :P
13:36:02 Join Mathiasdm [0] (
13:36:32pixelmapondlife: I believe it's initialised if it finds a voice file present and I think it starts speaking without a reboot if I enable the options (because the voice files are present)
13:37:03pondlifepixelma: I think it just needs to stop and restart playback to allocate, but I may be remembering wrongly.
13:37:56pixelmaanyways, I enabled voice menus just now and the reported buffer size did not change a bit...
13:38:03pondlifeUrgh, having enabled voice menus during playback, I get a nasty distorted voice
13:38:17pondlifeRecognisable - the correct clips, but distorted
13:39:23pixelmaseems it's not the logf build now. I remember now, had to reconfigure after the language buffer commit and didn't chose logf again
13:39:25pondlifeI suspect RBUtil's voice building - a reboot doesn't help
13:39:29 Quit Soap_ (Read error: 110 (Connection timed out))
13:41:42*pondlife hunts a paperclip
13:42:02pondlifeHard lock on shutting down
13:43:39*preglow vanishes
13:47:14 Quit BigBambi ("CGI:IRC (EOF)")
13:50:32 Join nplus [0] (
13:51:11 Join MethoS-- [0] (
13:53:16 Quit jcollie ("Ex-Chat")
13:53:36 Join petrus [0] (n=petrus@
13:53:48 Nick petrus is now known as DjLaurenz (n=petrus@
13:54:32DjLaurenzis there someone helps me to port a mp3 player for rockbox?
13:54:50DjLaurenzi'm not a developer but i want to try to port a T-Logic for rockbox
13:55:25GodEaternot keen to repeat that with such a large file
13:56:14linuxstbDjLaurenz: Porting Rockbox to a new device is a _lot_ of work - it's only done by developers who own such devices and are interested in using Rockbox on them.
13:56:38DjLaurenzcan't i help in any kind?
13:57:00linuxstbDjLaurenz: Help who?
13:57:35linuxstbSo far, only you want to see Rockbox ported to those devices... You need to find owners of those devices and encourage them to port Rockbox.
13:57:50DjLaurenzi understand...
13:58:33DjLaurenzi want to port it because t-logic is a good mp3 player but its original firmware is very very bugged
13:59:10DjLaurenz(sorry for my english, i'm italian :D )
14:01:58DjLaurenzthank you, bye bye
14:02:02 Part DjLaurenz ("Konversation terminated!")
14:02:15amiconnGodEater, linuxstb: I already mentioned the possible use of test_disk for testing USB speed
14:02:23 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
14:02:51amiconnCompile a sim with a modified test_disk that creates its test folder in a sub-folder instead of the root
14:03:23amiconnThen create a symlink with that name in the sim's archos/ directorsy
14:04:38amiconnLinusN: Bookmarking and resume seem to be buggy, but only on swcodec
14:04:54linuxstbamiconn: Yes, I remember you saying that. A standalone app would be nicer though...
14:06:52LambdaCalculus37GodEater: Around?
14:08:29 Quit MethoS- (Read error: 110 (Connection timed out))
14:10:49 Join agm3nt [0] (
14:14:53pondlifeYou can make the sim use any folder as its root with the −−root <dir> command-line option; might that be helpful?
14:14:55 Quit XavierGr ()
14:16:33 Join defukt [0] (n=WTFOMGBB@
14:17:01amiconnpondlife: That would mean everything would have to reside on the target's disk, including the sim's .rockbox folder
14:17:14amiconnBut that would work, even in a sub-folder
14:17:30amiconnOr just install the whole sim in a path on the target disk
14:21:07 Quit Febs (Read error: 110 (Connection timed out))
14:31:23 Join jcollie [0] (
14:31:44 Join midkay_ [0] (
14:42:45***Saving seen data "./dancer.seen"
14:43:37 Part LinusN
14:49:14 Quit midkay (Read error: 110 (Connection timed out))
14:52:48 Nick fxb__ is now known as fxb (
15:00:05 Join perrikwp [0] (i=98216e29@gateway/web/ajax/
15:00:20 Quit BlakeJohnson87 (Read error: 110 (Connection timed out))
15:02:15 Join kugel [0] (n=kugel@unaffiliated/kugel)
15:02:32 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
15:04:04 Join Davide-NYC [0] (
15:04:12 Join Gold_Bug [0] (
15:04:13 Nick Gold_Bug is now known as Gold (
15:04:27 Nick Gold is now known as GoldBug (
15:04:28Davide-NYCpetur: ping
15:04:43nplusis this my fault?
15:04:56nplussansa e200, r16365
15:04:58peturDavide-NYC: pang
15:05:06Horschtsuper pang 3
15:05:34Davide-NYCStupid question: do I need to reflash my RAM image after extracting my patched build?
15:06:07Davide-NYCThis is in reference to CF card HDD replacement in H1x0 devices (for the logs)
15:06:34peturDavide-NYC: no idea, I never used a flashed RB install because I have an h300
15:07:26Davide-NYCah, so then you are having a stable experience without having flashed the rockbox.iriver file to RAM.
15:10:54peturI flashed a modified bootloader to get past the ata -80
15:11:01 Nick fxb is now known as fxb__ (
15:11:13peturfor the rest it is a normal disk install
15:11:49Davide-NYCpetur: would the contents of a debug screen from after an ata -80 error help in solving this?
15:12:25Davide-NYC:( what can I do to help?
15:12:43Davide-NYCI *LOVE* the boot time. It's incredible. Maybe 2 seconds (if that)
15:13:04peturI started a support query with A-Data (doing this from work did yield an initial response but nothing helpfull)
15:13:09markunhi Davide-NYC!
15:13:20LambdaCalculus37Hey Davide-NYC!
15:13:36peturHey LambdaCalculus37!
15:13:42markunhi petur!
15:13:48LambdaCalculus37Hey petur!
15:13:50peturmust be friday :)
15:14:17petur1:45 until company drink :D
15:14:24LambdaCalculus37Davide-NYC: Have you got a spare HDD to CF adapter on hand? I'd like to see if this mod will work on other targets.
15:14:28Davide-NYChell no, it's 9:15am and I gotta get going very soon. (hehe)
15:14:42LambdaCalculus37In this weather, Davide-NYC? It's awful out there. :)
15:14:55Davide-NYCI work two blocks from my house.
15:15:01LambdaCalculus37Lucky you. :)
15:15:03peturLambdaCalculus37: the adapters are _very_ cheap on ebay
15:15:39LambdaCalculus37I'll get some and see if we can even try modding an iPod.
15:15:53*petur wonders what kind of length measurement 'blocks' is
15:16:29Davide-NYCpetur: I'll be running the patched build today. Is there anything in particular you can tell me to look out for or pay special attention to?
15:16:53Davide-NYCOK thanks. Later all !
15:16:57peturI thought it was very unstable?
15:17:11LambdaCalculus37Davide-NYC: See ya!
15:17:13Davide-NYCNot very. Occaisional intermitent freezing
15:17:29Davide-NYChard to pinpoint what fails
15:17:43Davide-NYC(when fails?)
15:17:45 Quit Davide-NYC ("ChatZilla 0.9.81 [Firefox]")
15:17:48peturpermanent lockup or continues after 30 seconds?
15:19:19 Join Febs [0] (n=chatzill@
15:22:08 Join n1s [0] (
15:24:13 Join magmaniac [0] (
15:28:33 Quit Mathiasdm ("Yuuw!")
15:32:11 Join Moize [0] (
15:32:17 Join perrikwp|class [0] (i=982136a8@gateway/web/ajax/
15:35:11GoldBuggreetings I have a question
15:36:02 Join nicktastique [0] (n=nick@unaffiliated/nicktastic)
15:36:25markunGoldBug: welcome
15:36:45GoldBugI was curious
15:37:11GoldBugi am thinking about getting an Ipod but i HATE ituned
15:37:25GoldBugbut i don't see any versions of rockbox for the 160 gb
15:37:39LambdaCalculus37And you won't for quite a long tiime, it seems.
15:37:48LambdaCalculus37No one is working on porting Rockbox to the iPod Classic.
15:37:49markunI don't think anyone is seriously working on a port for the ipod classic
15:38:04GoldBugsad why is that?
15:38:11 Join Evilnick_work [0] (n=0c140464@gateway/web/cgi-irc/
15:38:12linuxstbGoldBug: The 160GB is the so-called "Classic' ipod, which is completely different to other ipods, and hence requires a new Rockbox port which no-one is working on.
15:38:13RamlaAren't there a lot of replacements for itunes, if that's the only problem
15:38:40GoldBugnot that i am aware of
15:38:42markunEvilnick_work: good morning
15:38:52GoldBugonce you go ipod your pretty much locked into it
15:38:55LambdaCalculus37Evilnick_work: Morning, Nick!
15:38:57 Join mf0102 [0] (n=michi@
15:39:03markunGoldBug: are you a programmer?
15:39:13 Quit Moize ("Ex-Chat")
15:39:14GoldBugand I have been seriously considering getting an ipod for a while just to put on rockbox
15:39:15GodEaterthere are all sorts of iTunes alternatives
15:39:28GoldBugi am a coder not a programmer
15:39:35LambdaCalculus37GoldBug: Close enough.
15:39:46GoldBugwell lemme clarify
15:40:08GoldBugi can code on a simple lever
15:40:21GoldBugi have never gone down to the level of assembly or working on drivers
15:40:31markunThere are some things you can do to help with a port, even if you are not a programmer:
15:40:39GoldBugi am just a hobbist coding
15:40:47LambdaCalculus37So are the majority of us.
15:40:49linuxstbGoldBug: You sound like lots of people before they started getting involved with Rockbox...
15:40:49GodEatermost of which he can't do if he doesn't already own the ipod markun ;)
15:41:25GoldBugwell the thing i am most concerned about it bricking a 350 dollar ipod if i screw up
15:41:45GoldBugi just liked the way that rockbox looks it's sexy^^
15:42:01LambdaCalculus37GoldBug: If you have the money to burn, why not do this: buy a supported iPod, like the iPod video.
15:42:05LambdaCalculus37Then get a Classic to help do the porting effort.
15:42:21linuxstbGoldBug: If you want to run Rockbox, you should buy a player which is already supported by Rockbox. Sadly that doesn't include any 160GB devices, unless you include the very old Archos devices that use 2.5" hard drives.
15:42:43GoldBugwell that is the thing i don't NEED 160 gb
15:42:51GoldBugi just WANT 160 gb =p
15:43:16linuxstbDon't we all? ;)
15:43:26 Quit GodEater ("Ex-Chat")
15:43:34*moos is waiting for his new toshiba gigabeat S with a future 120GB disc ;)
15:43:36LambdaCalculus37GoldBug: You can buy an 80GB video and upgrade the drive later on.
15:43:46Evilnick_workmarkun: Hi there
15:43:49 Join GodEater [50] (n=bryan@rockbox/staff/GodEater)
15:44:11GoldBugyou can actually replace the hdd and it doesn't freak out>?
15:44:19linuxstbmoos: Are 120GB drives easily available currently?
15:44:20GoldBugi spose you are replaceing the OS
15:44:23Evilnick_workLambdaCalculus37: Hey!
15:44:38mooslinuxstb: yup it seems so, around 200$
15:44:46linuxstbGoldBug: Even without Rockbox, most (if not all) mp3 players can have their hard drives replaced.
15:45:10*GoldBug is mostly ignorant on the subject
15:45:44linuxstbGoldBug: See this page for some info on the topic -
15:45:49*Evilnick_work has a nice 120GB iPod Video thanks to being able to change HDD
15:46:11GoldBugwhat kinda drives do they run? laptop hdds?
15:46:20GodEater1.8" laptop drives
15:46:25linuxstbGoldBug: See the link I just posted...
15:46:25GodEaterfor the most part
15:46:32GoldBugyea i just loaded it
15:46:41 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
15:46:46LambdaCalculus37Some, like the iPod mini, use 1" Microdrives.
15:48:15*GoldBug is SERIOUSLY reconsidering
15:48:33GoldBugmaybe i should go the 80 gb vid ipod route
15:49:14*amiconn wonders what happened to that ata50-zif adapter offer...
15:49:26Evilnick_workIf you're going for capacity then the cheapest way of doing it would be to get the 120GB drive and then get a second-hand iPod video (60 or 80) that has a bad HDD
15:50:01GoldBugnah i really don't need it that much
15:50:18GoldBugand now that i think about it most of the video that i would wanna put on it is subtitled
15:50:26Evilnick_workI've currently got the best part of 80GB free!
15:50:38GoldBugand it would be hard as hell to read subs on something that is a 2.5" screen
15:50:49LambdaCalculus37Evilnick_work: How much did you spend to get your iPod video?
15:51:55GoldBugno the majority of the reasoning behind wanting rockbox is that i don't have to convert any of my files
15:52:25Evilnick_workLambdaCalculus37: $74.01
15:52:37Evilnick_work+ shipping
15:52:43LambdaCalculus37GoldBug: See? Saves you lots of money to get a used video.
15:52:46Evilnick_work+$200+ for the 120GB drive
15:53:07LambdaCalculus37Overall still cheaper than a $349 160GB Classic.
15:53:09Evilnick_workIt was cheaper 'cause the stock HDD wasn't working.
15:53:13linuxstbGoldBug: Are you aware that Rockbox doesn't have support for lots of video codecs? It just plays a very specific type of mpeg-1/2. Rockbox's video playback on the ipod video is also poor (it doesn't use the video decoding hardware on that device).
15:53:31Evilnick_workYeah, I suppose so. And as it uses Rockbox, it's FAR superior to the Classic 8)
15:53:41GoldBugi was mostly worried about audio codecs
15:54:13linuxstbGoldBug: Audio codecs we do... (unless the files are encrypted with DRM).
15:54:26*GoldBug hisses at DRM
15:54:44GoldBugyea that I know
15:55:48 Join Llorean [0] (n=DarkkOne@
15:55:52alienbiker99does anybody know how to use serndfirm on windows?
15:57:31GodEaterI don't think it works on windows yet does it ?
15:57:48linuxstbI thought jhMikeS had it working on Windows (or something similar).
15:58:21 Quit GoldBug ("ChatZilla 0.9.81 [Firefox]")
15:59:00amiconnlinuxstb: Btw, did you check your build server so that it can be reenabled?
15:59:20GodEateramiconn's not been reading the logs again
15:59:27alienbiker99ok, i cant seem to get the driver to work. is it easier to use it in linux?
15:59:28pixelmahe answered that yesterday...
15:59:37linuxstbamiconn: LinusN has now re-enabled it.
15:59:58GodEateralienbiker99: it's an absolute doddle in linux
16:00:03linuxstbalienbiker99: Yes, it "just works" in Linux...
16:00:19*n1s 3rds
16:01:25alienbiker99hm ok, i think when i tried to compile it, i ended up with an error
16:02:06n1sypu need libmtp-dev
16:04:23 Join jgarvey [0] (
16:05:09alienbiker99ah ok, ill get it and try to compile again
16:06:41GodEater <−−- good news for ipod owners!
16:07:36 Quit Evilnick_work ("CGI:IRC")
16:07:43n1swow, the OF isn't very fast
16:07:53 Quit MethoS-- (Read error: 110 (Connection timed out))
16:08:13LambdaCalculus37It's faster than the Emergency Disk Mode now!
16:08:43*LambdaCalculus37 buys GodEater a beer
16:08:59LloreanGodEater: Which iPod is that on?
16:09:04GodEater5.5G 80GB
16:09:08LloreanI was about to say
16:09:11GodEaterI'm just editing my wiki page now
16:09:12LloreanThose times have to be a video
16:09:22LloreanBecause they're WAY too fast for Nano's emergency disk mode.
16:09:32LloreanWell, the ratio of E-disk to OF-disk rather.
16:09:34LambdaCalculus37GodEater: Want me to try it on my 5.5G 30GB?
16:09:46LloreanIf that's where we're at, I'm in favour of enabling it for Nanos today. =P
16:10:13GodEaterLambdaCalculus37: by all means
16:10:16GodEaterthe more tests the better
16:10:33LambdaCalculus37Right. I'll compile a build at home and test.
16:11:01*LambdaCalculus37 hopes he gets to leave work early today because of the nasty as hell snowstorm
16:11:51 Quit kkurbjun ("Leaving.")
16:11:54 Join toffe82 [0] (
16:13:51GodEaterMy test file is 2 Megabytes of "Rockbox!\n" over and over so I could check it had actually appeared in the firmware partition easily
16:14:13GodEaterLlorean: I don't suppose your nano is out of a box somewhere is it ?
16:14:16LambdaCalculus37I'm going to try this one: time dd if=/dev/sdb1 of=/dev/null bs=32k count=1000
16:14:54LambdaCalculus37GodEater: Just to make sure, where do I define #USE_ROCKBOX_USB?
16:15:02GodEaterin the Makefile
16:15:13GodEaterfind the line which reads EXTRA_DEFINES=
16:15:19GodEaterand add "-DUSE_ROCKBOX_USB"
16:15:37GodEateras linuxstb pointed out though - the write tests are the important ones for people
16:15:38 Quit toffe82 (Read error: 104 (Connection reset by peer))
16:15:47GodEatersince it's the use case that'll happen most often
16:16:16 Join remotechief_311 [0] (
16:16:26LambdaCalculus37Okay, got it.
16:16:52GodEatermy next test is writing to the filesystem
16:16:52LambdaCalculus37Could I use your test file for when I get to the test?
16:17:09LloreanGodEater: No, it isn't. But I'll probably have it found by Monday at the latest, if all goes as intended.
16:17:10rasher(Or you can define it in the firmware/export/config-*.h file)
16:17:19rasherIf I remember that path correctly
16:17:35*GodEater thinks the Makefile route is easiest
16:17:45 Join toffe82 [0] (
16:17:58rasherYou have to redo that when you reconfigure
16:20:19linuxstbGodEater: I'm not sure about the "bs=1" in your tests - IIUC, that means that dd will be writing one byte at a time, and I'm not sure how the kernel would deal with that...
16:21:09GodEaterI can rerun it without
16:21:56LambdaCalculus37linuxstb: That means that I shouldn't really use the bs=1 argument then when I dd the test file.
16:21:59GodEaterif you think it makes that much difference
16:22:19GodEaterI think that means you should leave out the count= too
16:22:21linuxstbI would expect it to be much faster with a larger bs
16:22:54linuxstbe.g. ipodpatcher does writes to the device in 32KB chunks, which was much faster than even 512 byte chunks.
16:23:02GodEaterwould you suggest giving a value for bs=, or would you let dd decide on it's own ?
16:23:06 Nick midkay_ is now known as midkay (
16:23:17linuxstbafaik, it just defaults to 512.
16:25:05GodEaterwell I can test with both 512 and 32k if you want
16:25:11 Join DaCapn [0] (
16:27:05GodEaterI just used the bs=1 because it seems to use the value in there for calculating the offset to start writing at
16:27:22GodEaterso if I use a bs=512, I need to rejig the seek= parameter too
16:27:31linuxstbYes, "skip", "seek" and "count" all use the block size specified by bs
16:27:43GodEaterwhich is a pain in the arse
16:28:07linuxstb(or obs...)
16:28:19*linuxstb hands GodEater a calculator
16:28:35GodEaterI have one thanks
16:30:23*LambdaCalculus37 is going for coffee
16:31:11 Part agm3nt
16:32:00 Join stewball`ghost [0] (n=WTFOMGBB@
16:36:54 Quit defukt (Read error: 110 (Connection timed out))
16:36:55 Quit toffe82 (Read error: 104 (Connection reset by peer))
16:37:03 Part perrikwp|class
16:37:45 Join toffe82 [0] (
16:37:49 Join SimonAdameit [0] (
16:40:37alienbiker99im trying to install this libmtp package in vmware and now its asking me to update the linux kernel, i cant seem to get it to update, any suggestions
16:40:57GodEaterI doubt it'll work from vmware anyway
16:41:12GodEaterthe USB support from guest->host in vmware is rudimentary at best
16:41:32alienbiker99crap. i just had to delete my linux partition the other day
16:42:21n1sYou should have deleted the win partition :P
16:42:21 Quit toffe82 (Read error: 104 (Connection reset by peer))
16:42:49***Saving seen data "./dancer.seen"
16:43:38 Join toffe82 [0] (
16:43:50GodEaterthe larger block size tests are less encouraging
16:46:09SimonAdameitI have installed the current build of rockbox for Ipod mini 4th generation according to the manual installation instruction. But when booting I get a "Bad checksum". Here is the bootscreen:
16:47:00LambdaCalculus37Umm... Apple never released a 4th generation of the mini.
16:47:02linuxstbSimonAdameit: There isn't a "4th generation" - are you confusing that with a 4GB model?
16:47:30SimonAdameitI have a 4GB model
16:47:49linuxstbThen you have either a first or second generation mini. Does it say "4GB" on the back?
16:48:15LambdaCalculus37Then get this:
16:48:22toffe82anybody want to have a look to this firmware :
16:48:32toffe82or help me to start with
16:48:36SimonAdameitLambdaCalculus37: thank you, i will try it
16:48:37linuxstbSimonAdameit: Then it's a "second generation mini"
16:48:51LloreanGodEater: I wish I had my Nano around. =/
16:49:45toffe82it is based on a pp5022
16:50:40LambdaCalculus37toffe82: Wonder if it shares hardware with other targets...
16:50:46linuxstbtoffe82: I was just about to tell you that.. It looks a bit like the mi4 files.
16:51:36linuxstbtoffe82: Yes, it's an "mi4" file. So we should be able to easily run code on that device.
16:51:41toffe82I will receive it next week and will open it to see the hardware, just wanted to know if it looks like something existing
16:53:40SimonAdameitThank you very much. It works now using the right model.
16:53:41 Quit toffe82 (Read error: 104 (Connection reset by peer))
16:54:45 Join toffe82 [0] (
16:54:52LambdaCalculus37SimonAdameit: You're welcome. :)
16:57:31 Join defukt [0] (n=WTFOMGBB@
16:58:08toffe82linuxstb: what should I do to start to work on it
16:58:20 Part SimonAdameit
16:59:28linuxstbtoffe82: Firstly find out the firmware upgrade process, and what recovery procedures there are...
17:00:28linuxstbPlus the obvious task of disassembling it and identifying the components.
17:00:42linuxstb(physically disassembling)
17:01:31toffe82I look at some stuff on the philips site, and there is a program to update and "repair" also there is a rest button, so it seems impossible to brick it
17:01:35GodEatertoffe82's favourite bit
17:01:49 Quit stewball`ghost (Read error: 110 (Connection timed out))
17:02:08 Quit Axio (Read error: 110 (Connection timed out))
17:02:09linuxstbtoffe82: Does "reset" just reboot it, or do more?
17:02:15toffe82I change from toshiba to philips :)
17:02:52linuxstbAlso, does the "update and repair" program require any special button combinations to be pressed on the device (to enter a recovery mode)
17:03:21 Quit petur ("connection reset by beer - very nice beer")
17:04:07 Join SSnake [0] (
17:04:27SSnakehi all
17:04:36LambdaCalculus37Hmm... from the Philips FAQ for the GoGear, it looks like the device is MTP only.
17:04:44toffe82from the doc :
17:04:46toffe82In the unlikely event that your player hangs, do not panic. Insert a small pin or other sharp
17:04:47toffe82object into the reset hole located at the back of the player. Hold until the player shuts down
17:05:30SSnakewhere can i find lcd refresh frequencies?
17:06:33Nico_Prockbox got a mention on planet ubuntu :)
17:06:39linuxstbLambdaCalculus37: Good job we now have a USB driver for portalplayer then... ;)
17:06:49SSnakedo we?
17:06:54SSnakegreat to know!
17:07:07LambdaCalculus37We are oh so very close to officially including it.
17:07:23linuxstbSSnake: It's still not perfect, but it's functional. Main issue is speed and some bugs with the integration into Rockbox.
17:08:01SSnaketoo bad i'm not so great rev eng and not good electronics knowledge
17:08:21SSnakei think it''d really interesting to develop such thing
17:08:35SSnake*it'd be
17:09:45SSnakeanyway...nobody can help me?
17:09:48 Join hi [0] (n=5956093e@gateway/web/cgi-irc/
17:10:14SSnakethe point is i'm limiting refreshing routine to the lcd maximum FPS refresh rate
17:10:35SSnaketo some some cpu cicle and optimize things
17:10:39linuxstbDo you mean the internal refresh frequency, or the speed Rockbox can refresh the LCD?
17:11:02SSnakeinternal, the speed the lcd can support
17:11:13hiis ipod 160Gb supported by rockbox?
17:11:15SSnakejust to avoid unuseful refresch
17:11:21linuxstbhi: No.
17:11:48 Quit hi (Client Quit)
17:12:03SSnakelinux don't you know any source to look for in which this problem is considered?
17:12:12GodEaterthat's like the 3rd person to ask that today
17:12:15*GodEater wonders what's going on
17:13:07 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
17:13:26linuxstbSSnake: No.
17:13:31 Join Domonoky [0] (n=Domonoky@
17:14:22 Quit toffe82 (Read error: 104 (Connection reset by peer))
17:14:40SSnakedamn! it would be very useful to avoid wasting cpu time in unnedeed refresh cicles
17:14:57linuxstbHow often are you updating the LCD at the moment/
17:15:44 Join toffe82 [0] (
17:15:51SSnakemmm honestly i don't know: i'm just porting a game from linux to rockbox and i didn't calculate the exact timing refresh
17:16:08 Join LarryFenske [0] (n=d811893a@gateway/web/cgi-irc/
17:16:20SSnakeprobably around 50 or 100hz
17:16:29linuxstbYou're much more likely to be struggling to refresh at a high frequency than refreshing too much...
17:17:27SSnakemy quest is to avoid a function call that refresh the lcd if this is not needed
17:18:40LloreanDepending on the game, there's no reason not to pick a speed even lower than the refresh rate of the screen.
17:18:51LloreanSomething arbitrary like 20, 25, or 30hz.
17:19:51LarryFenskeHi. I have been using Rockbox for a while and have recently started using USE_ROCKBOX_USB on my Sansa E200. I would like to add my measurements to the UsbPerformance twiki page and need permission to do so.
17:20:02LambdaCalculus37LarryFenske: Sure. What name?
17:20:18LarryFenskeLarryFenske (I wanted to make it simple!)
17:20:26LambdaCalculus37One moment.
17:21:07LambdaCalculus37Welcome to the wiki!
17:21:30SSnake....just to let developers know: the nearest neihbour resize routine works really well!
17:22:08SSnakei use it in my game and it's capable of resizing a 320x240 buffer to 220x176 realtime in the game
17:22:13SSnakeon a h320 iriver
17:22:52LarryFenskeWhile I'm here (assuming this is the appropriate venue), the UsbPerformance page has a table with one entry and one of the columns is for "Emergency Disk Mode". I can't find a reference to such mode in the Sansa. Is it similar to recovery mode or manufacturing mode?
17:22:56SSnakei adapted to just to support "keep aspect ratio", but it's quite the same code you can find in bmp.c
17:23:08LarryFenskeLambdaCalculus37, thanks for adding me.
17:23:10GodEaterLarryFenske: there's no such mode on the sansa
17:23:12 Join stewball`ghost [0] (n=WTFOMGBB@
17:23:18GodEaterso an N/A entry would be appropriate there
17:23:21LarryFenskeThanks, GodEater.
17:23:34GodEaterI'm just finalising the filesystem writing performance tables
17:23:40GodEaterso you won't be able to edit the page just yet
17:24:02LarryFenskeOK, I'll wait.
17:24:28*gevaerts points out that bs=1 doesn't match any real-world usage at all.
17:25:18GodEater <−− filesystem tests
17:25:27GodEatergevaerts: hence the later additional tests
17:26:11LambdaCalculus37Rockbox UMS peformed better with 1024 byte chunks.
17:26:25GodEaterLarryFenske: ok - all yours
17:26:37 Quit defukt (Read error: 110 (Connection timed out))
17:26:42linuxstbSSnake: What are you resizing (and how often)?
17:26:54gevaertsCan you also test 128k blocks ? At least in linux that will be the commn case (for the normal case, sync will be different)
17:27:09GodEatergevaerts: I was looking for someone to tell me what the default would be
17:27:12GodEaterI'll do those too
17:27:25GodEaterdid you notice we beat the EDM in the filesystem tests so far
17:27:44SSnake@linuxstb: a bitmab buffer of fb_data array 320x240 pixels to fbdata 220x176 pixels
17:28:08LambdaCalculus37GodEater: We're kicking ass with the USB test speeds. :)
17:28:36GodEaterwell - not quite
17:29:39 Quit courtc (Remote closed the connection)
17:29:40GodEaterbut that's definitely an improvement for Rockbox using ipod video owners who don't use the OF to do their transfer work
17:29:42 Join courtc [0] (
17:29:55gevaertsGodEater: true, but both speeds are horrible. Probably the sync mode kills it all. Maybe measure with time "cp sb_write_test_file /mnt/ipod/Data/usb_test;umount /mnt/ipod) " ?
17:29:57 Quit z35 ("Leaving")
17:30:03pondlifeGodEater: So, would you now consider using Rockbox UMS?
17:30:24pondlifeI mean, to save a reboot to copy a couple of files over, not to do a complete refresh...
17:31:01GodEatergevaerts: I'll do that too
17:31:05linuxstbSSnake: Yes, but _what_ is in that buffer, and how often do you resize it?
17:31:10GodEaterprobably not today though :)
17:31:13GodEaterpondlife: yes I'd use it
17:31:22pondlifeI'd be interested in comparing that to something with a hardware bridge too - you don't have any Archos or Iriver models lying around do you?
17:31:27gevaertsGodEater: of course, you'll have to calculate the speed by hand then...
17:31:52GodEatergevaerts: I'm sure I'll cope ;)
17:32:05 Join MethoS- [0] (
17:32:36 Join defukt [0] (n=WTFOMGBB@
17:32:59SSnake@linux: it's fb_data! RGB packed values. Honestly i don't know how often: i wanted to make some test with a sort of fps limiter but it's impossible without knowing the lcd refresh rate
17:33:25GodEaterthats with 128K blocks
17:33:28GodEaterwe still beat EDM
17:33:41SSnake@linux: (well not impossible, but it's inaccurate and not portable to other targets)
17:34:33*LambdaCalculus37 is missing out on all the fun of testing USB because his laptop's home :(
17:34:44*GodEater wouldn't describe it as "fun"
17:35:26*GodEater notices LarryFenske has buggered up the wiki page
17:35:50GodEaterno, I tell a lie
17:35:53GodEaterit was HarrisonFu
17:35:55GodEaterwhoever that is
17:36:52linuxstbSSnake: Do you mean that the game renders the display at 320x240, and then each lcd update does a resize?
17:37:07LambdaCalculus37Jeez, the page is a right mess now!
17:37:18GodEaterthank god for version history
17:37:19SSnake@linux: yes,exactly
17:38:08SSnakethe game runs just a bit slow without cpu_boost, but absolutely ok with it
17:38:38krazykitGodEater, that'd be harry tu on the tracker, or countrymonkey in irc.
17:38:48linuxstbSSnake: Well, you're wasting a lot of CPU cycles...
17:38:59pondlife...and battery.
17:39:06LloreanYeah, if you want to worry about efficiency, I'd tackle rendering in native resolutions before I worried about skipping frames.
17:39:09SSnakeof course i know but it's the most portable way
17:39:33linuxstbNot with some Makefile magic and compile-time processing...
17:39:47SSnakeif not, i'd have to totally reenigneering the game
17:40:18linuxstbYes, which is normally the best thing to do if you want it to be efficient.
17:40:30linuxstbDo you know how much RAM it's using?
17:40:31SSnakeyeah you 're absoultely right :>
17:40:54SSnakei'm investigating into this stuff
17:41:21 Join Arathis [0] (
17:41:42*gevaerts thinks that SSnake has a point. Get it working quickly, commit it, wait for some efficiency expert to like your game. Problem solved...
17:41:53SSnakeafter a while if u have play enabled the games stops due to not enough ram
17:42:16LarryFenskeGodEater: It feels good to be redeemed.
17:42:26SSnakebut i have to investigate better on memory handling
17:42:43GodEaterLarryFenske: sorry, I leapt to conclusions :(
17:42:45linuxstbSSnake: Removing that 320x240 buffer would be a nice start...
17:42:49*Llorean wonders what game is being worked on.
17:42:56Lloreanlinuxstb: Sounds like a leak though if it runs out over time.
17:43:24GodEaterright - no more performance tests for me for today
17:43:31GodEaterbesides, the page is still locked
17:43:48 Quit MethoS- (Remote closed the connection)
17:44:05 Join Axio [0] (
17:44:29 Quit stewball`ghost (Read error: 110 (Connection timed out))
17:44:34 Join seablue [0] (
17:44:40LarryFenskeGodEater, maybe I'll try tomorrow. Gotta go. Thanks.
17:44:52kugelSSnake: what game where you talking about?
17:45:23SSnakeit's a surprise....have to wait two weeks i think
17:46:26SSnake@linux yeah i have to change data type....2 bytes for every pixel is really too much for the game
17:46:32gevaertsCurrently the usb stack is optimised for transfers larger than 32k. Below that the double buffering won't work and disk speed will start to get involved. The larger the better...
17:46:32 Quit LarryFenske ("CGI:IRC (EOF)")
17:47:22kugelSSnake: does your irc client not support nick tab completition?
17:47:46linuxstbSSnake: Where do you get memory from? The main audio buffer or the plugin buffer?
17:47:59SSnakeno i'm using opera integrated irc client
17:48:30kugelSSnake: You didn't tell me yet of which game you're talking about
17:48:31SSnakeplugin buffer of course, but if there's no play it uses all audio buffer
17:48:42 Join x1jmp [0] (
17:48:52SSnake@kugel: i told you it' a surprise :>
17:49:19kugelSSnake: Oh that was directed to me
17:49:44*gevaerts remembers SSnake talking about an old school platform game
17:49:44kugelmaybe I could help you
17:50:31linuxstbSSnake: Why does it use the audio buffer only if there's no playback? Either it needs the memory or it doesn't...
17:50:35SSnakenot now: i'm doing a university work for my thesis. Later i will commit all to svn
17:51:54SSnake@linux: right question. First i wanted to check memory usage. Second i'm planning to add sound only if playback is stopped (audiobuffer available)
17:51:55kugelSSnake: You have commit access? Who are you then? I can't spot your nick on the irc nick table
17:52:10SSnakeno i haven't
17:52:43SSnakedata sound is something around 3MB
17:52:45kugelI wonder how you're gonna commit it then :)
17:53:14GodEaterkugel: maybe he has his own repo ;)
17:53:21 Quit defukt (Read error: 110 (Connection timed out))
17:53:22kugel:) it's time to come back to last exam....anyone want to hel me in signal and sistem exam?
17:54:38SSnake...i guess nobody wants :> ok seeya later.
17:55:02kugelI'm off now cya
17:55:04 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
17:55:05 Part SSnake
17:55:14*gevaerts has a speed measurement to make people happy :
17:55:42GodEaterthat's just a read right ?
17:55:54dionoeawith the original firmware ?
17:56:07*GodEater shows dionoea the door
17:56:19 Join einhirn [0] (
17:56:25*dionoea opens the door and leaves
17:56:27GodEaterdon't let it hit you on the way out ;)
17:56:40gevaerts also has a write. And this is rockbox
17:56:55GodEaternice going ;)
17:57:04*gevaerts found out what made high speed not work...
17:57:24*GodEater wins!!!!
17:57:28 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
17:57:32saratogais it PP being crap
17:57:43*Llorean throws a party.
17:57:54 Quit mf0102 ("Verlassend")
17:58:09*GodEater thanks gevaerts for rendering all his work today on performance figures for full speed a complete and utter waste of time :)
17:58:26linuxstbGodEater: What else are Friday afternoons for?
17:58:32GodEaterbeer in the pub
17:58:35 Quit Horscht ("We don't make mistakes, we just have happy little accidents")
17:58:44linuxstbDon't you at least need to be pretending to work?
17:58:52GodEaterthe boss is in Japan
17:59:00GodEaterand the boss's second in command is in Worcester
17:59:03GodEaterso no - not really ;)
17:59:25gevaertsBasically the problem was that the controller got enabled, reset, diabled again, and then on connect pin detect it got enabled again (but no reset) before being initialized. Now I keep it enabled.
17:59:43gevaertsenabled == DEV_EN, reset == DEV_RS
17:59:47*GodEater awaits the new code in svn with glee
18:00:19*gevaerts reminds GodEater that some people actually will use full speed. Not everyone has a new(ish) pc
18:00:23*pixelma is a little late in the \o/ party but joins in anyways :)
18:00:35GodEaterI wonder if there's wiki markup for a big cross across the whole page :)
18:00:42GodEatervery true gevaerts, very true
18:00:45GodEaterI stand corrected
18:01:08jhMikeSgevearts: congrats \o/ /o\ \o/ /o\
18:01:13*linuxstb was about to do some speed tests, but will now wait impatiently
18:01:14x1jmpI’m wondering why the battery in my h120 discharges that fast, now I found out that the player needs ~300mA at 3.8V (no backlight, no disk activity)
18:01:20GodEaterjhMikeS: show off ;)
18:01:35x1jmpDoes someone have a clue, what might be wrong with it?
18:01:41GodEaterjhMikeS: it's your job to get all this working on the S now :)
18:01:54n1sx1jmp: how did you measure?
18:02:26x1jmpn1s: connected a multimeter between a battery wire
18:03:00jhMikeSGodEater: I just know it will all go off without a hitch :)
18:03:42GodEaterand on that note - I have to go meet a man about a pint
18:03:48GodEaterhave a lovely weekend everyone
18:04:20jhMikeSthanks...and have designated driver :)
18:04:22x1jmpI just bought a new 2200mAh battery, but it still plays for about 7h
18:05:34n1sx1jmp: 300 mA is quite a bit too much...
18:06:06n1shave you enabled the optical out?
18:06:52x1jmpn1s: no, ATM I’m testing with default settings
18:06:57*linuxstb gets around 965KB/s reading and 1.0MB/s with the USB driver in SVN (dd to the raw device on an ipod Color)
18:07:54n1sx1jmp: is the disk spinning a lot?
18:08:01Domonokyx1jmp: you could check is some chip is getting hot (perhaps power regulator) (only a guess)
18:08:46*gevaerts recommends linuxstb to svn update
18:08:49 Join Mathiasdm [0] (
18:09:19gevaertsI just committed the high-speed fix (nothing else yet, I'm still busy reorganizing code)
18:10:22*x1jmp searches for the power regulator ...
18:11:20jhMikeSgevaerts: so...RB is going a bit faster than retailos (4.6 vs. 4.5 iirc)?
18:11:29 Join ender` [0] (
18:11:55gevaertsjhMikeS: I think that's within measurement error. We are slower on write though.
18:12:07*gevaerts waits for someone to optimize the SD driver
18:12:30 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:12:36pixelmathe frontpage script still doesn't know about gevaerts's full name...
18:12:39amiconngevaerts: The controller should not be kept enabled when there is no usb connection, as it will suck power this way
18:12:43jhMikeSI tried to optimize SD writes but the speed always trashed the fat for some reason
18:13:30gevaertsamiconn: my fix doesn't keep it enabled. The first test did, but that's a different story. It was just missing DEV_RS bits when re-enabling
18:13:42x1jmpDomonoky: the Coldfire is a bit warm, but not very much though
18:14:12jhMikeSgevaerts: it's always some little thing like that the causes so much trouble for so long :\
18:14:46linuxstbgevaerts: Read is now 3.2MB/s on my ipod Color.
18:15:02x1jmpAccidently I had connected ~9V to the DC jack for some seconds, I’m not sure if that could have done some damage to the device
18:15:07gevaertsjhMikeS: it's a good thing that you wanted the threads sorted out. That lead me to this particular code, otherwise I might not have looked at it
18:15:13Domonokyx1jmp: thats it !
18:15:24Domonokythen its surely the power converter..
18:15:39linuxstbgevaerts: "bs" seems to make very little difference - bs=512 takes 13.313s, 128KB is 12.793 seconds.
18:16:02Domonokythe hxx0 players are really easy to damage with a wrong voltage /polarity..
18:16:05jhMikeSgevaerts: glad to be of help! :)
18:16:18 Quit n1s ()
18:16:19 Quit toffe82 (Read error: 104 (Connection reset by peer))
18:16:28BigBambigevaerts: woohoo! congrats :)
18:16:31Domonokyx1jmp: you are lucky that it still works.. :-)
18:16:36x1jmpDomonoky: Are they replacable?
18:16:58gevaertslinuxstb: It did make a difference on full speed, because of turnaround times. On high speed the disk is very much the limiting factor, not what you actually do on usb. And bs won't matter much on a PIO disk driver
18:17:09linuxstbgevaerts: Those were reading tests btw
18:17:18markunGodEater: is anyone working on making the USB code work on the Gigabeat S?
18:17:25gevaertslinuxstb: any idea about EDS and/or OF ?
18:17:25amiconnx1jmp: Depends on what is broken, which is difficult to tell
18:17:33linuxstbgevaerts: I'll test those next...
18:17:48Domonokyx1jmp: if it really is only the powerconverter it should be possible to solder a new one in... perhaps ask LinusN, has has done a few hxx0 repairs i think..
18:17:50linuxstbgevaerts: I'm just doing a write test now, but that seems to be taking a long time...
18:17:50 Join toffe82 [0] (
18:18:05BigBambimarkun: I believe Nico_P had a go but didn't get too far
18:18:16amiconngevaerts: I don't expect we'll get usb as fast as the values reported for G5 OF
18:18:30x1jmpwhere is the power converter on the PCB?
18:18:38amiconnOur ata driver doesn't use DMA, and afaik nobody knows how to do that
18:18:45amiconn(on PP)
18:19:05*gevaerts goes to have dinner now
18:19:20linuxstbgevaerts: Hmm, writing (bs=512) was 283KB/s, but bs=128k gave 5.4MB/s
18:19:31Domonokyx1jmp: take look at the wiki, it has shematics and pics i think..
18:20:01x1jmpHaven’t found it there yet
18:20:27jhMikeSI think that the S's SDMA interface is also similar. I'm getting the feeling that looking at freescale stuff might shed light on other PP stuff (just a hunch).
18:23:35 Quit toffe82 (Read error: 104 (Connection reset by peer))
18:24:39 Join hannesd_ [0] (
18:25:44 Join toffe82 [0] (
18:28:50 Quit hannesd (Connection timed out)
18:28:50 Nick hannesd_ is now known as hannesd (
18:31:39 Quit toffe82 (Read error: 104 (Connection reset by peer))
18:31:40 Join toffe82_ [0] (
18:31:48 Nick toffe82_ is now known as toffe82 (
18:32:04jhMikeSwhoa...big oops...I mean SDHC, not SDMA :0
18:33:30amiconngevaerts: I wonder why the 2 USB devices are handled separately instead of at once
18:33:37linuxstbInitial comparison with Rockbox high-speed UMS vs RetailOS on my ipod color - a 310MB directory of FLACs took 67 seconds to copy in Rockbox, and 29 seconds in RetailOS
18:33:56linuxstb(mount ; cp ; umount)
18:34:25 Join MajorC [0] (
18:34:37amiconnI would expect the following stuff to work: DEV_EN |= (DEV_USB0|DEV_USB1); DEV_RS &= ~(DEV_USB0|DEV_USB1);
18:35:31BigBambi6.5 MB/s read on my e260!
18:37:40linuxstbEMD seems identical to RetailOS on my Ipod Color - 29s for the same directory.
18:38:14Llorean If I recall, we've only had reports of EMD being slower in Video and Nano.
18:38:30linuxstbYes, but I've never actually tested it before...
18:39:29 Join Soap_ [50] (n=Soap@rockbox/staff/soap)
18:41:08 Quit MajorC ()
18:42:09x1jmpIs that LTC3405 the power regulator in the h120?
18:42:50***Saving seen data "./dancer.seen"
18:44:16toffe82jhMikeS: did you see the link I gave yesterday for the register of the mc13783 ?
18:44:59 Quit magmaniac ("Leaving")
18:45:01jhMikeStoffe82: yeah. noone has tried contacting freescale about more detailed info?
18:45:45toffe82not as I know
18:47:37 Join waldo [0] (
18:47:59 Quit remotechief_311 (Read error: 110 (Connection timed out))
18:49:49*barrywardell wonders if it's going to be time to release a new sansa bootloader soon with the new usb code
18:50:22linuxstbbarrywardell: Do you mean with the USB stack, or just the "boot into Rockbox, not OF" change?
18:50:37barrywardelleither one
18:50:39linuxstbBTW, have you tried the Rockbox stack on a Mac recently?
18:50:52barrywardellalthough it would be nice to get rid of the sansa bootloader too
18:51:09barrywardelliirc recovery mode is the main thing holding that back
18:51:29linuxstbYes, but if our bootloader has a full UMS mode, then I think that's better...
18:51:32barrywardelllinuxstb: I tried it before the latest commit and still don't get a disk appearing. trying it again now
18:51:54gevaertsamiconn: no idea. Maybe because it made experimenting (enabling or disabling just one) easier at an early stage ?
18:52:02barrywardelllinuxstb: I agree. UMS mode in the RB bootloader would be ideal
18:52:26linuxstbbarrywardell: Although for it to be useful, a PC-side application to recover would probably be needed (rbutil?)
18:52:28 Join webguest65 [0] (n=4db40aa8@gateway/web/cgi-irc/
18:52:45 Quit webguest65 (Client Quit)
18:52:53pondlifeDomonoky: In case you saw my comments from earlier, alleging that RBUtil produced distorted voice files, the problem was that somehow my RBUtil Speex volume was set to 10.00... ;)
18:53:14pondlifeSetting it back to 1 worked nicely
18:53:39 Join DerDome [0] (
18:53:40 Join Nikolaus [0] (n=4db40aa8@gateway/web/cgi-irc/
18:54:08barrywardelllinuxstb: wouldn't sansapatcher do the job?
18:54:31 Join Benjamin_L [0] (
18:55:00 Join countrymonkey [0] (n=3dd50610@gateway/web/cgi-irc/
18:55:26linuxstbbarrywardell: Possibly. Although I'm wondering what would need recovering... If the Rockbox bootloader loads, then the user just needs to reformat and/or copy across a new
18:55:28 Join petur [50] (n=petur@rockbox/developer/petur)
18:55:51barrywardelllinuxstb: just tested on my mac. still no drive appearing, but USB Prober reports "Rockbox media player" and disk utility finds "Rockbox Internal Storage Media"
18:55:58gevaertsI think that to get a real idea of UMS performance issues, measurements should include test_disk output as well (with boosted cpu).
18:56:25gevaertsbarrywardell: can you do a logf build with logf enabled for usb_storage.c and report what that says ?
18:56:50BigBambiwell, 3min35 for rb and 2min31 for OF to copy 286 MB on my e260 :)
18:56:54*BigBambi is quite hqppy
18:56:54linuxstbgevaerts: Do you think "dd" to a partition device is enough, or should we be copying files?
18:57:27barrywardellgevaerts: sure. just a second...
18:57:28 Quit Nikolaus (Client Quit)
18:58:15*amiconn will test against his Amiga's USB stack later
18:59:03 Part pondlife
18:59:07gevaertslinuxstb: I think either method will do. The problem with copying files is that you either need to unmount or sync to flush everything, so it's harder to get a reliable measurement.
18:59:21barrywardelllinuxstb: the only thing that would ever need recovering would be the bootloader itself, in which case e200tool would be required anyway
18:59:24countrymonkeyI feel that wallisertitsch should be removed from the zips; it is very close to the 50% mark that shows that it is very outdated. Rasher's page shows it hasn't been outdated for 3 years.
18:59:57*gevaerts thinks that the actual bugs are more important than performace tweaks now.
19:00:03Domonokypondlife: jeah there was a bug in the settings of rbspeex in rbutil.. and as it remembers settings, the fix only helps if you reset the rb settings :-)
19:00:09rashercountrymonkey: we do not exclude any languages
19:00:49countrymonkeyIIRC afrikaans, turkçe and a few more were excluded (or was that a patch).
19:01:01pixelmacountrymonkey: you were the one asking about the japanese charset of the Archos Player, right? Did you see amiconn's answer the next morning in the logs?
19:01:12rashercountrymonkey: That was in the patch. It was not enabled when committed.
19:01:35gevaertsI should have committed with "some speed improvements" in the message, and kept silent here, just to see how quick people would notice
19:01:36countrymonkeyYes it was me, but I didn't see it. I will check it. Thanks.
19:04:15rashergevaerts: a late congratulations
19:05:08barrywardellgevaerts: I also just got this on linux:
19:05:22*jhMikeS is confusled by the delta is his last commit
19:05:24linuxstbbarrywardell: Have you tried the USB on your H10?
19:06:19 Quit Soap_ (Read error: 110 (Connection timed out))
19:06:22gevaertsbarrywardell: that's interesing... Why does the mac keep reading the same data over and over again ?
19:06:24barrywardelllinuxstb: no, not since gevaerts started working on it. It's at home so I can try later and report back when I do
19:06:33 Quit perrikwp (" ajax IRC Client")
19:06:37countrymonkeyThanks for the answer; I know I cannot do it now. (katakana is the system of foreign language words)
19:06:53waldohi guys , first off ... rockbox rocks, my sansa is a miracle machine now that it plays all formats, syncs to and does replaygain ... HOWEVER, i went skiing and fell on my player (bragging mode : 180 crossed :P ). I lost one channel . So i opened it up, it seems that the connector for the tip of the jack has come off the PCB .. i tried to solder it back on but found that the solder island was janked of the pcb so i have nothing to
19:06:53waldosolder onto, so my question ... can someone tell me where i can find diagrams so i can find a point to put a wire to ? or can someone find one ... i know this isn't realy the place ... but i'm kinda desperate ; if i send it back to sandisk (still in waranty) i have to remove rockbox and i'll prolly get a V2 replacement wich would kinda suck ...
19:07:04amiconnlinuxstb, barrywardell: It's working nicely on small H10 (fullspeed, and apart of the known bugs of course)
19:07:47amiconnJust that this test revealed that the H10's microdrive is to blame for the slow usb access, and not the OF :/
19:07:51BigBambiwaldo: you may (should?) find pics on the wiki that may help, I'm not sure
19:08:11jhMikeSwaldo: I'm not aware of them having a problem with rockbox...they did want it ported afterall (no secret there)
19:08:16barrywardellgevaerts: it does detect that there are two partitions (and gets the size right), so it's making some progress
19:08:46linuxstbamiconn: We need to implement the serial number for the H10s and m:robe 100. It's currently only done for ipods (we use the same serial number as the OF - read from the flash), and the Sansas (reading a serial number from the AS3514).
19:08:57countrymonkeyI have my doubts about you getting a v2 replacement. (I have a curse on me which always makes me send things in for repair; I never get a newer model)
19:08:57barrywardellgevaerts: recovery mode doesn't work on my mac either, so it must be some weird apple peculiarity
19:09:10amiconngevaerts: Do you think the slow mounting on Windows could be due to repeated stuff as well?
19:09:20barrywardellamiconn: hopefully tthe big H10's drive will be faster
19:09:34BigBambicountrymonkey: Please do not give advice based on absolutley nothing
19:09:44gevaertsbarrywardell: it reads sectors 600-607 lots of times, without rockbox having or reporting any error there.
19:09:48waldobtw any luck on porting to the V2 ? cause i'd rather send it back then having a sansa that is patched up like that.
19:09:58amiconnlinuxstb: Yeah, true. Do you know how to find the serial number that an usb device reports to the pc?
19:10:09linuxstbamiconn: Only on linux - "lspci -v"
19:10:15amiconn(then that number could be used to search a H10 rom dump)
19:10:18linuxstbI mean lsusb...
19:10:32amiconnOr OF...
19:10:46amiconnNah, nmust be rom, I think
19:10:53barrywardellgevaerts: I can dump those sectors on linux if it's any help?
19:11:11gevaertsamiconn: could be. The "easy" thing about windows is that I have a windows machine at work next to a usb tracer, so I can see exactly what it's doing. I can't do that with a mac unfortunately (none available there...)
19:11:12jhMikeSI thought the S/N thing was resolved already
19:11:42linuxstbjhMikeS: Only on ipods and Sansas afaik
19:11:52 Quit countrymonkey ("CGI:IRC (EOF)")
19:13:08gevaertsbarrywardell: maybe. I'd like to finish the integration cleanup first though.
19:13:42saratogaanyone object to me setting cabbie as the default theme when I get chance?
19:13:48saratogai've more or less got that worked out
19:14:03barrywardellgevaerts: <- they're the sectors 600-607 for when you get around to it
19:14:04gevaertslinuxstb: there's still the potential issue that on my c250 our serial number does not match the OF's serial number at all. It might still be unique, but I have no idea if that's really true
19:14:37 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
19:14:48Lloreansaratoga: Might want to post a patch for it, so people can get a look at it and see if there's any problems you've missed. But I'm all for having it default to it soon
19:14:50jhMikeSnoone tried reading i2c rom yet?
19:15:13linuxstbjhMikeS: Does the AS3514 datasheet mention a serial number ?
19:15:17saratogaLlorean: I posted a patch last week thats essentially ready to go
19:15:23LloreanAh, missed that then.
19:15:26saratogaall thats needed is minor change to the build script
19:15:30LloreanWas it a new task, or in one of the old threads?
19:15:45jhMikeSlinuxstb: it has the 120-bit fuse array which can be programmed with anything
19:15:52gevaertslinuxstb: do you have any objections to making all serial numbers the same length ? We are deviating from the OF anyway now, and it would cut down the #ifdef forest a bit
19:15:58 Quit einhirn (Read error: 104 (Connection reset by peer))
19:16:37linuxstbjhMikeS: Ah, OK.
19:16:54gevaertsbarrywardell: That's at sector 600 ? It looks like a partition table.
19:17:07linuxstbgevaerts: No, Zagor convinced me that there was no need to match the OF's output.
19:18:12gevaertslinuxstb: OK. I'll remove the "duplicate" declaration then. Any preference on whether the ipod s/n should be left- or righ-aligned in the longer string ?
19:18:13 Quit toffe82 (Read error: 104 (Connection reset by peer))
19:18:22dan_awaldo: I've just started doing a bit of work on the Sansa v2 - I'm planning to get a C200 soon. AFAIK, though, we don't yet know how to write the firmware, so it could be a very long way off
19:18:37 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:18:48linuxstbgevaerts: Doesn't the UMS spec talk about 12 digits that must be unique?
19:19:37Horschtmake high speed work?
19:19:42Horschtcan I have your e-babies?
19:19:43 Quit Mathiasdm ("Yuuw!")
19:19:45 Join toffe82_ [0] (
19:19:53 Nick toffe82_ is now known as toffe82 (
19:19:56linuxstbgevaerts: Or maybe truncate the Sansa string? 16 characters would seem enough...
19:20:02gevaertslinuxstb: indeed. the last digits, so that answers my question.
19:20:51linuxstbAlthough it might be hard to know which digits to drop... So maybe best to keep all 32 and pad the ipod string.
19:21:01 Quit Arathis ("Bye, bye")
19:21:47gevaertslinuxstb: we could use a cryptographic hash of the serial number... Best keep it as-is I guess
19:22:36barrywardellgevaerts: oops, forgot to skip (or is it seek?)
19:22:48linuxstbskip for input, seek for output
19:24:09 Quit toffe82 (Read error: 104 (Connection reset by peer))
19:25:31 Join Simsgy [0] (
19:25:34gevaertsbarrywardell: wild idea, can you try changing the USB_VENDOR_ID and USB_PRODUCT_ID to something else ? Maybe macs assume a specific behaviour if they see an ipod ?
19:25:36SimsgyHey guys
19:25:47 Join toffe82_ [0] (
19:25:55 Nick toffe82_ is now known as toffe82 (
19:26:01SimsgyI was wondering... Is there ANY support for Nano 3g yet? (I know there isn't just wanted to make sure :( )
19:26:27SimsgyBecause the firmware is un-breakable?
19:26:33*gevaerts really wants a rockbox usb vendor id. Anyone got $2000 to spare ?
19:27:24barrywardellgevaerts: this is on my sansa, but I'll give it a shot anyway
19:27:31Horschtarent't there "generic" ones?
19:27:44gevaertsbarrywardell: then it won't help I fear.
19:28:16LloreanSimsgy: Because nobody is working on it.
19:28:22Domonoky$2000 is heavy for a usb vendor id... we need a sponsor.. :-)
19:28:33gevaertsHorscht: no. And companies can't give you a spare product id according to the usb-if conditions (although they ould grant an exception I guess)
19:29:11SimsgyLlorean: Why
19:29:26gevaertsDomonoky: I think that any $2000 can probably be spent better elsewhere. This is just a purist cleanlyness thing
19:29:31BigBambiSimsgy: What kind of question is that?
19:29:38saratogai don't think a vender id is all that desirable
19:29:40BigBambiIf people want it, they need to work on it
19:29:41Horschtbecause no one with interest in porting it has a 3g nano
19:29:49BigBambiSimsgy: They aren't
19:29:56SimsgyWell, thanks
19:29:57saratogaprobably better to use the OF's, that way we can potientially still support things like itunes or OF's software
19:30:07Horschtand obviously, the people that own a 3g nano don't care about porting it
19:30:08 Part Simsgy
19:30:51 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
19:34:05barrywardellgevaerts: no luck :(
19:34:06 Quit toffe82 (Read error: 104 (Connection reset by peer))
19:34:40 Join toffe82_ [0] (
19:34:49 Nick toffe82_ is now known as toffe82 (
19:37:13 Quit linuxstb (Nick collision from services.)
19:37:14 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
19:38:02linuxstbgevaerts: There are companies that sell individual vendor/product IDs very cheaply
19:38:08gevaertsbarrywardell: Can you try at full speed, i.e. revert to just before my last fix ? There are only reads in there, and some of them seem to be timing out.
19:38:19gevaertslinuxstb: really ? where ?
19:38:59Horschtthank you, jhMikeS, for the temporary fix on the spinlock issue
19:39:41gevaertsbarrywardell: in case you didn't notice, I'm just throwing out wild ideas. I really have no idea right now as to what the problem is.
19:40:06rasherI wonder if this spinlock thing was related to the awful skipping performance on the e200 when shuffling many songs
19:41:08linuxstbHmm, the link has died...
19:41:27Horschtworks here, linuxstb
19:41:32 Join dandin1 [0] (
19:41:41Horschtoh, no wait. it doesnÄt
19:42:19midgeybarrywardell: my mac says it was "not able to enumerate a device" and will keep retrying according to dmesg when i try to plug in my e200
19:42:50 Join low_light [0] (i=c730180b@gateway/web/ajax/
19:43:03linuxstbThe issue of whether we want our own IDs was discussed on 12 October 2007, starting at 15.11.06
19:45:25*Domonoky just sees 10 USB-PIDs for 25 ... so much cheaper... :-)
19:45:25 Quit toffe82 (Read error: 104 (Connection reset by peer))
19:45:44 Join toffe82_ [0] (
19:45:52 Nick toffe82_ is now known as toffe82 (
19:46:26linuxstbDomonoky: Yes, I think I've found the same -
19:46:56linuxstbAlthough it says "Pending a discussion with the USB organisation I have suspended the sale of USB PIDs" - so maybe the USB org have put a stop to these resellers...
19:47:52Horschtdoesn't the USB org offer a very low volume license or something?
19:47:56*low_light reports that he has buttons working on mrobe 100
19:48:12linuxstblow_light: Have you tried the USB driver on your mrobe?
19:48:28gevaertsSo I was right, USB-IF doesn't like people sharing their vendor id.
19:48:47linuxstblow_light: Congratulations btw ;)
19:49:24 Quit Benjamin_L ("Leaving")
19:49:36low_lightlinuxstb: thanks, still needs some tweaking
19:49:48Horschtgevaerts, yeah obviously. They want to get more money.
19:51:18low_lightlinuxstb: I haven't tried the usb driver yet
19:52:10linuxstblow_light: Something to add to your to-do list would be to find out the USB serial number the OF uses, and try to find out where that's stored on the device (so Rockbox can use it).
19:52:26linuxstb(in Linux, use "lsusb -v" and look for iSerial)
19:52:30gevaertsThe main problems I see with using the OF ids are : (a) we _will_ behave differently in some corner cases (hopefully by being more correct), and host drivers use id-based black- and whitelists to enable/disable certain bug workarounds. (b) we may present different device classes than the OF
19:53:08low_lightlinuxstb: do you have the tatung elio bootloader? (not the OF)
19:53:26linuxstblow_light: I should have, yes.
19:53:37low_lightI mean the OF bootloader
19:54:03gevaertsAnd I'm not really interested in having just one product id. I think real advantages to our own id only start appearing if we can allocate one per hardware platform (like differentiating between 32 and 64 MB ipods), and unfortunately that's not economically viable right now...
19:54:40low_lightlinuxstb: I want to see if i can dig out the lcd code for you ;)
19:55:21linuxstblow_light: That would be nice...
19:55:31linuxstb(see PM...)
19:55:52alienbiker99is there a list of rockbox users bidding on ebay for mp3 player? so that one user doesnt outbid the other?
19:56:32rasheralienbiker99: surely people would be bidding the same max amount anyway?
19:56:35gevaertsalienbiker99: how do you decide which one is allowed to bid today ?
19:56:52Horscht"I saw it first!"
19:56:59BigBambialienbiker99: I neither see how it would work, nor think it should
19:57:00LambdaCalculus37"Dibs on this one!"
19:57:15alienbiker99dunno just asking. ive seen channels that have a list set up like that, but there was a lost more people wanting it than just the community
19:57:24Horscht"you already got 7 players, i only have 5!"
19:57:57LambdaCalculus37"Tough! *I* get to have the Coldfire target this time!"
19:58:03gevaerts"I need just this one to complete my collection of first editions of rockbox-supported players!"
19:58:14LambdaCalculus37alienbiker99: This is an example of what would happen.
19:58:20barrywardellgevaerts: sorry, I was away for a bit there. just tried full speed and still no luck :(
19:58:21alienbiker99haha yeah i know
19:58:25Horscht"If you bid on this one, I'll slit your throat!"
19:58:42*gevaerts thinks that Horscht might win this argument
19:58:49LambdaCalculus37"I'll go Feyd-Rautha on your pasty ass!"
19:59:01*LambdaCalculus37 has had his fun :)
19:59:03Horschti don't want to win in the intertubes, gevaerts
19:59:38LambdaCalculus37alienbiker99: A prime example of why a list like that is a terrible, terrible idea.
19:59:43gevaertsbarrywardell: do you have a usb hub available ? Or are you using one ? Maybe changing bus topology helps.
19:59:52LambdaCalculus37We're going to end up killing each other.
20:00:05gevaertsLambdaCalculus37: wouldn't that solve the problem ?
20:00:16barrywardellmidgey: did you put the #define in your config?
20:00:27LambdaCalculus37gevaerts: But who gets to enjoy the DAP in the end, when everyone's congealing on the floor?
20:00:31Horschtit would, but this channel would be "dead" quiet, gevaerts
20:00:45*gevaerts hopes there would be one survivor...
20:00:47Horscht(pun intended"
20:01:14midgeybarrywardell: I added #define USE_ROCKBOX_USB to config-e200.h
20:01:40barrywardellgevaerts: I'm on a laptop, so only two ports to try. both same problem
20:03:44gevaertsbarrywardell: if you manage to try with a hub, please do. It shouldn't make a difference (since we can't control the wire-side of things, it should behave the same as the OF), but you never know...
20:06:35linuxstbsaratoga: Looking at your patch for FS #8443, it seems to include unrelated changed to do with backdrops - were those intended?
20:06:35 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:06:44 Join toffe82_ [0] (
20:06:52 Nick toffe82_ is now known as toffe82 (
20:10:34barrywardellmidgey: weird. is that with current svn?
20:11:26midgeyi'll take a look at it later tonight; i'm actually on my way out for spring break
20:11:26barrywardellgevaerts: will do. in case it's any help, here's what dmesg says on my mac:
20:11:59midgeyim running 10.4.11 Tiger on a Macbook
20:12:25barrywardellI'm 10.5.2 Leopard on Macbook. Maybe that's the difference
20:13:06 Quit midgey ()
20:14:21 Join cars [0] (n=451de957@gateway/web/cgi-irc/
20:14:22saratogalinuxstb: yes, pixelma complained that my previous cabbiev2 commits caused various color only theme options to be written on grayscale targets
20:15:43 Join hannesd_ [0] (
20:17:22 Quit hannesd (
20:17:23 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:18:00 Nick hannesd_ is now known as hannesd (
20:18:10 Join toffe82_ [0] (
20:18:18 Nick toffe82_ is now known as toffe82 (
20:18:20gevaertsbarrywardell: that dmesg output looks like an OS bug to me. Hard to be sure though.
20:19:33 Join spiorf [0] (n=spiorf@
20:20:26linuxstbsaratoga: Shouldn't that be a separate patch then?
20:20:59barrywardellgevaerts: could be. although it doesn't happen with the OF USB mode
20:22:12barrywardellgtg. I'll report back if I find anything more of interest
20:22:47 Quit barrywardell ("Konversation terminated!")
20:23:12 Quit linuxstb ("Leaving")
20:24:19 Quit cars ("CGI:IRC (Ping timeout)")
20:24:21 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
20:26:43 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:26:44 Join toffe82_ [0] (
20:26:52 Nick toffe82_ is now known as toffe82 (
20:30:44 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:30:46 Join toffe82_ [0] (
20:30:54 Nick toffe82_ is now known as toffe82 (
20:33:04 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:34:38 Join toffe82_ [0] (
20:34:46 Nick toffe82_ is now known as toffe82 (
20:35:52amiconngevaerts: Something seems to be wrong with highspeed mode
20:36:16amiconnI can mount my mini g2, but file transfer is *way slower* than with the fullspeed version
20:36:48 Quit DaCapn (Read error: 110 (Connection timed out))
20:38:09gevaertsamiconn: fun
20:38:09 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:38:16amiconnHrrm, and now it even stopped working
20:38:44linuxstbamiconn: Is that on Windows?
20:40:39 Join toffe82_ [0] (
20:40:47 Nick toffe82_ is now known as toffe82 (
20:41:00gevaertsamiconn: there is something at least somewhat wrong here too : I get a device reset just after scanning the partitions. It doesn't seem to harm, but maybe it does on windows
20:41:32gevaertsamiconn: Note that this is on not on exact svn
20:41:35 Quit gevaerts (Remote closed the connection)
20:41:48amiconngevaerts: I compared unzipping a rockbox build using 7-zip on windows. With diskmode it takes 16 seconds, with fullspeed rockbox it took 35 seconds
20:42:27linuxstbThat's about the speed difference I got with diskmode vs highspeed Rockbox on my Color
20:42:30amiconnBut with highspeed rockbox the predicted extraction time was > 10 minutes, and then it stopped ~ 1/3 way through and complained that the system could not find a file
20:42:52***Saving seen data "./dancer.seen"
20:43:50 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
20:44:09*gevaerts hates accidentaly typing "sudo halt"
20:44:52*petur hands gevaerts a gui
20:45:54gevaertsamiconn: I'll test on windows soon.
20:46:58gevaertspetur: GUI's are worse. Look in the logs for the last time I plugged in a mouse.
20:48:35 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
20:48:44*amiconn thinks this windows problem and the macos problem are probably actually the same
20:49:36*gevaerts thinks that amiconn may be right
20:50:47 Join barrywardell [0] (n=barrywar@
20:51:20LambdaCalculus37I'm off now... see everyone later!
20:51:31 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
20:52:10 Join lee-qid [0] (
20:53:04*amiconn just tried mounting the thing on his Amiga
20:53:12*gevaerts is very afraid
20:53:17amiconn*PANIC* Writing on sector 0
20:53:38*amiconn wonders why AmigaOS wants to write to sector 0 on mount...
20:53:56gevaertsamiconn: I actually saw windows do that on my sansa as well a while ago
20:53:59*amiconn thinks we should disable sector 0 protection in the ata driver
20:56:26*gevaerts agrees. Why is it actually there ?
20:56:27 Quit toffe82 (Read error: 104 (Connection reset by peer))
20:56:38 Join toffe82_ [0] (
20:56:43amiconnAfaik it was added as a safety measure back when ata and filesystem write support were developed
20:56:46 Nick toffe82_ is now known as toffe82 (
20:57:27*gevaerts discovers md5sum mismatches on testing high-speed support
20:58:19gevaertsShould I disable high speed again in svn until we know what's going on ?
20:58:31amiconnThe (well, my) Amiga USB is fullspeed only, btw
20:58:54amiconnI doubt a classic Amiga would be fast enough to handle high speed anyway
20:58:59gevaertsamiconn: yes, but you have problems in windows with high speed, I get read corruption,...
20:59:53amiconnHeck, the CPU in that thing is slower than the one in the device I am connecting - and it's already the fastest CPU a classic Amiga can have
20:59:53 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:00:20*gevaerts suggests doing work on the DAP and playing music on the Amiga
21:01:06 Join stripwax [0] (
21:01:30 Quit homielowe (Read error: 110 (Connection timed out))
21:01:54 Join toffe82_ [0] (
21:02:02 Nick toffe82_ is now known as toffe82 (
21:05:40amiconnSo, anyone against ditching sector 0 write protection?
21:05:40 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:06:24amiconnThe MMC (Ondios) driver also has it, but neither the flash driver (iFP 7xx), nor the SD driver (Sansas)
21:06:38 Join toffe82_ [0] (
21:06:46 Nick toffe82_ is now known as toffe82 (
21:06:47peturdid anybody ever see that *panic* ?
21:06:55amiconnYes, me
21:06:55linuxstbIt doesn't matter on ipods - diskmode will start regardless of a valid sector 0 or not.
21:07:04amiconnSeveral times, this week
21:07:08peturI mean, when not testing usb ;)
21:07:11gevaertsDid it ever appear in normal usage ?
21:07:31 Quit hannesd (Remote closed the connection)
21:08:42*petur runs out of ideas regarding CF powermanagement setting failure
21:09:25*gevaerts is now actually working in two different source trees
21:09:59amiconngevaerts: I never saw it, but I came to the project when write support was already there
21:10:00linuxstbJust remember to compile the one you're working on...
21:11:00gevaertsamiconn: I think if noone remembers forum cries for help about it, it probably never occurred
21:11:09 Join ol_schoola [0] (
21:12:01amiconnThat was ~ 4 years ago, btw (coding wise, even a bit longer as a user)
21:14:19 Join MajorC [0] (
21:14:37 Join Domonoky_ [0] (n=Domonoky@
21:14:55 Join qwedsa [0] (
21:15:09amiconngevaerts: Something is odd on Amiga, btw: I get 2 (identical) disk icons
21:15:37 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
21:16:35amiconnIt's actually mounted twice
21:17:30gevaertsamiconn: interesting.
21:18:24amiconnThe serial number doesn't look right
21:18:51amiconnThe whole usb device info doesn't look right...
21:20:29gevaertsWhat's the policy on comitting trivial typo fixes in comments ?
21:22:07peturjust commit them?
21:23:18*gevaerts just did his first completely safe commit
21:23:42amiconngevaerts: Strange. The serial number sometimes matches the OF, and sometimes it is different
21:24:30amiconnThe difference, if it exists, is that certain bytes are 00 instead of the expected value - but not always the same bytes are affected
21:24:55gevaertsFor my usb integration work, I'd like to add HAVE_SOFTWARE_USB. Whether a target has it probably depends more on the chipset than anything else. Should I add it to all relevant config-*.h files, or just once to pp5020.h ?
21:25:09gevaertsamiconn: sounds like a cpu cache problem to me
21:25:17 Quit nicktastique ("Leaving")
21:25:20 Quit ol_schoola ()
21:25:38MajorCjust wanna thank you frank for an excellent work on the usb-support for portal players!
21:25:56*petur tried all all possible CF powermanagement levels, all refused :/
21:26:59 Join gibbon_ [0] (
21:27:08gibbon_good evening
21:27:40gevaertsamiconn: I'm nearly sure of that. The string descriptor related code has no cache-awareness magic at all. I'll add it now.
21:27:49gevaertsMajorC: you're welcome
21:29:32*gibbon_ bows in deep thankfulness before the work that has been done by the rockbox developers in the last few years... you really saved my ipod from me stopping to love it...
21:29:38 Join hannesd [0] (
21:30:10 Quit qwedsa_ (Read error: 110 (Connection timed out))
21:30:22gibbon_i just wanted to state that once again... after thinking twice, i think my question better goes to the community channel... sorry for bothering
21:31:50 Quit hannesd (Remote closed the connection)
21:32:02 Join hannesd [0] (
21:32:42 Quit Domonoky (Read error: 110 (Connection timed out))
21:34:27 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:36:44 Join toffe82_ [0] (
21:36:53 Nick toffe82_ is now known as toffe82 (
21:39:09amiconngevaerts: Btw, the USB serial number can be found in windows
21:39:19gevaertsamiconn: the usb serial descriptor bug should now be fixed in svn
21:39:43gevaertsamiconn: I guess somewhere in device manager/properties ?
21:40:24amiconnYes. You need to enable the detailed information mode though (via an environment variable)
21:40:46 Join stel [0] (n=stel@
21:41:05 Quit saratoga ("CGI:IRC")
21:41:20amiconnAh, no, that was something else
21:42:12 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:42:24amiconnIn the device manager, expand the USB sub-tree, click the respective usb storage device, then go to the Details tab
21:42:48amiconn(all translated from german windows, so the exact names might be different)
21:43:37amiconnIn the dropdown list, there is an item "bus relations" (??), which contains the serial # as part of the string
21:43:45 Join toffe82_ [0] (
21:43:54 Nick toffe82_ is now known as toffe82 (
21:44:59*gevaerts goes to a Windows machine to try
21:45:31 Join Soap_ [50] (n=Soap@rockbox/staff/soap)
21:45:59amiconnThe string is & separated, the serial number is the second one from the end (the last one is just "&0" here)
21:46:07 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
21:46:45gibbon_hmm... maybe i was wrong...
21:46:58 Quit bluebrother ("leaving")
21:47:03gevaertsFound it. Good to know, but I still find lsusb and /sys/bus/usb/devices/ easier
21:47:53gibbon_i wonder, if there is any way to distinguish a sansa c250 FM from the v2 version before buying ... maybe some relevant specs... i looked on the buyers guide and could not quite find what i looked for
21:48:52*Domonoky_ thinks the only sure way is the of version...
21:48:59 Quit axionix_ (Read error: 104 (Connection reset by peer))
21:49:39 Join axionix [0] (
21:50:31gevaertsgibbon_: _usually_ a v2 mentions audible support on the box, and v1 doesn't. My v1 did, so this rule doesn't always work. I don't know if they sell v2's without mentionning audible, but I wouldn't be surprised. As far as I know that's the only way to guess without opening the box.
21:50:41gibbon_Domonoky_: hmm... i was told in the neighbourhood that there is work in progress to support the later v2 models, too... so perhaps the v2 is a good start to help, even if it does not work in the first place ;)
21:50:45amiconnlinuxstb, gevaerts: The solution for the H10 iSerial is trivial - it doesn't have one
21:51:24gevaertsgibbon_: If you can open the box, usually v2's mention "v2" on the back, but same thing as with audible, sometimes they don't, and sometimes v1's are marked v2.
21:51:26gibbon_gevaerts: thats hard in an online shop :) but when it cones hard on hard, i can always send it back
21:51:38Domonoky_gibbon_: i think nobody works on sansa v2 .. so a very long wy for rockbox on it..
21:52:13gibbon_Domonoky_: oh, ok... then the "... i can always send it back" becomes very more likely
21:52:20amiconngevaerts: Btw, OF reports just '0', while rockbox reports '0000000000000000'
21:52:34gevaertsgibbon_: your best bet would indeed be to send it back if it's a v2.
21:52:43*gibbon_ nods...
21:53:03linuxstbamiconn: Seems spec-violation is common-place...
21:53:18linuxstbDoes the H10 just have one disk mode?
21:53:21gibbon_if its a v1, fm radio and recording work, i suppose?
21:53:44gevaertsDoes the H10's disk have a serial number ?
21:53:46amiconnRecording works, but I wouldn't recommend a sansa for recording
21:53:48BigBambiDo all European models have a radio?
21:53:55BigBambiof the c200s?
21:54:01 Quit barrywardell ()
21:54:45gevaertsgibbon_: if it mentions radio, it should have a radio. If it has one, rockbox supports it.
21:54:58gibbon_BigBambi: the shop calls the device "FM" explicitly... i found some without that in the name... so not all have it but there are some available
21:55:06gibbon_thanks again
21:55:16gevaertsgood luck
21:55:20amiconnlinuxstb: Btw, the Ondios also don't have a serial number - and they use a hardware USB-MMC bridge
21:55:41rashergibbon_: I think some that don't advertise FM, actually do have the fm chip
21:56:15gibbon_may be ...
21:57:01gibbon_the ones without are average 10 euros less in price than the ones advertised with... it seems that sandisk is even worse when it comes to model names than netgear... *shiver*
21:57:29gibbon_even the micro-sd works?
21:57:54linuxstbamiconn: I just looked at my Elio (my only non-ipod PP device), and that also reports "0" as the serial number....
21:58:09gevaertsgibbon_: micro-sd works, and rockbox supports sdhc on the v1, which the OF doesn't
21:58:34amiconnThe iriver also just has a rather generic iroduct string: "IRIVER Mass Storage Device"
21:58:42gevaertsamiconn: any chance to get a serial number from the disk ? That would be good enough
21:59:20amiconnAnd the OF reboots into ROM diskmode on iriver when detecting USB
21:59:21BigBambigibbon_: Yep, with the added bonus of sdhc support
21:59:35 Join gregzx [0] (
21:59:35amiconnSo only a single mode (@linuxstb)
21:59:43BigBambiUp to a theoretical 32 GB i believe (the OF stops at 2GB with standard SD)
22:00:01rasherIs that theoretical 32GB limit correct?
22:00:04gibbon_ok... i think there is a v2 series on the e series as well...
22:00:13BigBambirasher: Not sure, hence the I believe
22:00:19BigBambiBut I thought so
22:00:26BigBambigibbon_: Yes there is
22:01:31BigBambirasher: Apparently it isn't
22:01:38 Join magmaniac [0] (
22:02:09gibbon_then, i think, the c250 it is...
22:02:12rasherBigBambi: Based on what? I've seen conflicting reports, which is why I'm asking
22:02:22BigBambirasher: Wikipedia claims a theoretical maximum of 2048 GB
22:02:38BigBambiStandard wikipedia disclaimers
22:02:42BigBambi(from me)
22:02:53BigBambiHang on
22:03:01*BigBambi ought to search for micro-sdhc
22:03:04rasherBigBambi: But it was wikipedia who claimed the 32GB as well (at least that's where I read it), so I'm not sure what to believe
22:03:50BigBambirasher: I'm not sure either
22:04:28 Quit gregzx ("ChatZilla 0.9.81 [Firefox]")
22:04:29gibbon_lets just say "every sdhc on the market might work"
22:04:40gibbon_ymmv :)
22:04:52 Join gregzx [0] (
22:05:05rasherI haven't seen reports about any microsdhc cards not working
22:05:24*gevaerts goes shopping for a 2TB sdhc card
22:05:38BigBambiWould be nice wouldn't it
22:05:41amiconnDo you have a time machine? ;)
22:05:56*BigBambi needs to go shopping for a 120 GB to stick in the gigabeast
22:06:05magmaniacpetur: ping
22:06:32zerasher: 32G is the biggest they've made so far isn't it?
22:06:38peturmagmaniac: hi
22:06:53zerasher: i'd be surprised if they already hit the limit after the whole sdhc debacle
22:07:01gibbon_ze: they have? thats a really HUGE microsd then
22:07:27zegibbon_: oh for micro i think i just saw 12G get announced...
22:07:30magmaniacpetur: I was thinking about how we can detect whether our device has a disk or a CF card
22:07:30gibbon_bigger than my ipod color was before it passed away
22:07:39gibbon_ah, ok
22:07:40zegibbon_: but thats just form factor, has nothing to do with technical limits
22:07:51zesd is still sd
22:07:53*gibbon_ was in the micro sdhc context still
22:08:08gibbon_my bad
22:08:12magmaniacpetur: and the CF spec says that we could use the word 0 of the identify_info array for that
22:08:40 Join FOAD_ [0] (
22:08:41 Quit ch4os_ ("Lost terminal")
22:09:40peturmagmaniac: you mean value 848A?
22:10:03magmaniacpetur: yes, though it seems that a bunch of other values are also allowed.
22:10:11amiconnI think magmaniac Means bit 15 of that
22:10:19amiconn...and that actually sounds like an idea
22:10:27amiconnGuess what the Nano reports there?
22:10:29magmaniace.g. my card actually uses 044A
22:10:40petur044A here too
22:10:47amiconnAnd on Nano too...
22:11:09magmaniacthe big question is, what does a HD return there?
22:11:19amiconnBit 15 is set for an ata device
22:11:38amiconnAnd for CF it is obviously unset
22:12:26amiconnEek, the MK8007GAH seems to violate the specs here :(
22:13:12*gevaerts thinks the only specs that aren't violated are those that are not implemented at all
22:13:16 Join Arathis [0] (
22:13:32magmaniacI guess I'll swap CF and disk again...
22:14:22peturamiconn: shouldn't your 'dump identify to disk' be committed?
22:14:27amiconngevaerts: The disk does have a serial number
22:14:57amiconnThe question is whether it's a faked one
22:15:10amiconnAnyone else with a H10 6GB (not 5GB)?
22:15:17gibbon_thanks for the help so far...
22:15:28gibbon_the rest is up to me and mrs luck i guess
22:15:32petur5GB w/o radio here
22:15:52amiconnpetur: I don't think so, as it's quite special. In its current form it would also break all non-ata targets
22:16:02gibbon_and again... thanks for all the hard work :) there are people out there you make really happy
22:16:10 Join alison_oxo [0] (
22:16:14gibbon_see you later
22:16:51amiconnThe ST660211 also violates the specs and reports itself as non-ata ...
22:16:54peturamiconn: really? I think it is more usefull than 'dump rom' for instance
22:17:02 Part gibbon_ ("... and thanks for all the fish")
22:18:07 Quit nplus (Remote closed the connection)
22:18:10*magmaniac hates it when Torx screws drop from the table
22:18:16amiconnIt has another problem - you need to know whether the dump is from a little endian or a big endian target
22:19:06 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:21:32*gevaerts tells magmaniac to work on the floor
22:22:01magmaniacamiconn: can you tell us the full value of word 0 for these non-conforming devices?
22:25:20 Quit FOAD (Read error: 110 (Connection timed out))
22:25:20 Nick FOAD_ is now known as FOAD (
22:26:13amiconnMK8007GAH (iriver H180) reports 0x0040. ST660211 (iriver H10) reports 0x0c5a (both violating the specs). HMS360604D5CF00 (mini) reports 0x848a (conforming)
22:28:33magmaniacMK2004GAL (stock H120) reports 0x0040 too
22:29:33magmaniacand the CF spec even says that 0x0040 is one of the "preferred" alternate configuration values for CF cards!
22:30:00magmaniacwhich means there may be cards which report 0x0040.
22:31:03gevaertsI posted a new patch on flyspray that should fix the button press and screendump related bugs, and gets rid of the usb_core thread. I didn't commit it because it is not easily possible to disable it any more. I'd appreciate some review.
22:38:42 Part perrikwp
22:40:07magmaniacSo word 0 doesn't help us. There's of course the "model number" (ASCII string in words 27-46)... one could have a blacklist of models against which one can match this string, and if it's in the blacklist, activate things like the ATA -80 workaround. But that is so ugly...
22:42:44 Quit alison_oxo ("Ex-Chat")
22:42:56***Saving seen data "./dancer.seen"
22:43:01jhMikeSgevaerts: which thread is it using now?
22:45:41 Join stripwax [0] (
22:48:46jhMikeSgevaerts: nvm. I saw although cranking up the priority during connection or transfers might be wise.
22:49:56 Quit stel (Remote closed the connection)
22:49:56 Quit lee-qid (Read error: 110 (Connection timed out))
22:51:02 Quit seablue ("life, death, life, death")
22:51:46magmaniacpetur: when you first posted the idinfo.bin of your card for amiconn (a few days ago), I took the liberty of downloading it too... do I see correctly that the "model number" of your card is set to *all spaces*?
22:53:54gevaertsjhMikeS: the usb thread seemed the logical choice. Maybe we can just crank up the priority at all times. It only does something when a connection is detected, and spends the rest of its time in queue_wait, so a higher priority shouldn't harm anything
22:55:23jhMikeSif it almost never runs for any reason I agree
22:56:47stripwaxgevaerts - once again, huge props on the usb work! does the latest commit (fix caching issues) resolve the corruption on high speed or is the corruption coming from the controller somewhere?
22:57:01gevaertsjhMikeS: It only gets triggered by connection changes, and on ondio sp on mmc insert/remove
22:57:20peturmagmaniac: seems so
22:57:43gevaertsstripwax: it only fixes the usb serial number sometimes not being correct. I haven't looked into the high-speed problems yet
22:58:10pixelmagevaerts: on OndioFM it might do the same ;)
22:58:15magmaniacpetur: now that is very descriptive :( how is one supposed these things apart if almost everything violates spec?
22:58:22peturmagmaniac: then again, serial is not at all what is printed on the case
22:58:57magmaniacs/supposed/supposed to tell/
22:59:10peturmagmaniac: I've done some CF stuff at work in the past and I'm not surprised
22:59:12gevaertspixelma: right. I somehow missed that in my search for HAVE_MMC last time
22:59:46magmaniacpetur: ah, for me it's the first time that I go this close to hardware.
23:00:43*magmaniac notices that his CF card claims to support SMART
23:00:45stripwaxok., ta
23:01:04peturmagmaniac: actually, I only moved closer to hardware in the last years, I used to do windows stuff
23:03:13magmaniacany more ideas for HD-CF detection? One could take the failing set_features() call as an indication... :-/
23:03:37amiconnmagmaniac, petur: Specs violation seems to be common among flash memory cards.
23:04:32 Nick B--daW is now known as B00daW (
23:04:34 Quit ompaul (Client Quit)
23:04:47 Quit jhulst (Read error: 113 (No route to host))
23:04:54*petur is surprised nobody complained about SimpleTech cards, they really are simple...crap
23:05:09 Join linuxstb_ [0] (
23:05:15magmaniacI think it's preety widespread. At work I often stumble across hard disk and controller idiosyncracies... often solved by firmware upgrades
23:05:34amiconnIirc at least one of my MMCs also has an empty serial number, violating the MMC specs
23:06:12 Quit linuxstb (Nick collision from services.)
23:06:17 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
23:06:44peturfyi, the powermanagement setting failure gives as extended error (request sense) 0x1F
23:06:59peturnot very helpfull
23:07:42amiconnActually I did nor remember correctly, but I doubt that a serial number of 0x0000000a is correct...
23:07:44*petur silently hopes A-Data tech support will actually give a usefull answer
23:08:07*gevaerts thinks that amiconn might be an early adopter
23:08:24amiconnMy 2GB Magu MMC has that serial number...
23:08:53gevaertsMaybe you have the tenth one produced ?
23:09:28peturor better: prototype
23:10:12magmaniacRegarding the CF mod, I have a completely unrelated, practical question: what do you do to fix the adapter to the ATA connector? I think if I shake mine hard enough (or equivalently, take a long bus ride), the adapter might dislodge...
23:12:09peturhave not given it much thought yet
23:12:09 Quit gregzx ("ChatZilla 0.9.81 [Firefox]")
23:12:32magmaniacI suppose some kind of adhesive tape will do the trick
23:14:01peturamiconn: I assume the reason we do the reset in the control reg and not the config reg is to avoid re-configuring?
23:16:16 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
23:19:16 Join gregzx [0] (
23:20:13 Quit Domonoky_ ("Trillian (")
23:20:23 Join Domonoky [0] (n=Domonoky@
23:20:57amiconnpetur: I don't know...
23:21:15amiconngevaerts: iProduct and iSerial are totally broken now....
23:21:25amiconniManufacturer too
23:22:12gevaertsSo my latest fix wasn't...
23:23:07amiconnI get gibberish characters with (parts of) '0123456789ABCDEF' in it
23:23:12 Join BlakeJohnson86 [0] (
23:23:14 Join EspeonEefi [0] (i=espeonee@STRATTON-TWO-O-FIVE.MIT.EDU)
23:23:16peturamiconn: the reset we do retains settings, while the reset in the config reg is a real one... as we do not reconfigure, I guess it makes sense
23:23:33 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
23:24:09magmaniacum, does somebody know what "IORDY support" means exactly?
23:24:27petura hardware control signal
23:24:46magmaniacgo on :)
23:25:08gevaertsamiconn: fixed
23:25:44magmaniacI'm asking because I found the following in the CF spec "Support of IORDY is mandatory when PIO mode 3 or above is the current mode of operation"
23:25:50 Join ender [0] (i=krneki@
23:26:12magmaniacNow my card says it doesn't support IORDY, but it does support PIO 4.
23:26:29amiconnSame thing in the ata specs
23:26:42magmaniacAnd RB uses PIO 4 if available, right?
23:26:44amiconnJust another specs vioaltion, as usual
23:27:17magmaniacright, though I wonder if this one causes some of our problems?
23:27:23peturmagmaniac: I already tried PIO0, same issues
23:27:32gevaertsDoes anyone know what 'ehci_hcd 0000:00:1d.7: dev 4 ep1in scatterlist error -104/-121' means ?
23:27:47magmaniacpetur: thanks, OK
23:27:47amiconnOn h1x0, iordy isn't even connected, so even a disk that supports it couldn't use it
23:28:09magmaniacOK, you see I'm poking around in the dark...
23:29:18Bagdergevaerts: kernel/drivers/usb/core/message.c
23:29:45 Join z35 [0] (
23:30:14amiconngevaerts: Looking good now. Is the iSerial limited to hexadecimal characters, or is that just by convention?
23:30:38gevaertsBagder: so does that mean that I managed to trigger a ehci bug with rockbox ?
23:30:47amiconnI'm asking because the hdd serial is ascii, and often contains letters in addition to digits
23:31:12gevaertsamiconn: for mass storage it has to be hex. How long is the hdd serial ?
23:31:22Bagdergevaerts: I'm not really entitled to say, but there's a huge comment block around that error message in the kernel code that might mean more to you than it does to me...
23:31:23 Quit low_light (" ajax IRC Client")
23:31:46 Join greatgatsby [0] (n=jay@
23:31:50amiconn20 characters (plain ascii, not unicode), but I didn't find a hdd that uses all characters
23:32:03 Join ol_schoola [0] (
23:32:08gevaertsBagder: there's also this commented out BUG() macro... I think I'll probably need a tracer to see what the device actually does.
23:32:26pixelmaBagder: are you able to teach the front page gevaerts' name (the build table etc. already do)
23:32:27amiconnThe string is padded with spaces, but the actual serial "number" is sometimes right aligned within the field
23:32:37gevaertsamiconn: so we could make that a 40 characters hex serial I guess
23:32:41amiconn(i.e. left-padded with spaces)
23:36:20gevaertsamiconn: I guess that number is somewhere in the identify data ?
23:36:33amiconnWords 10..19
23:37:13amiconnTake care of endianness
23:38:14gevaertsCan I assume that any target that doesn't define HAVE_FLASH uses ata, or is there a specific one ?
23:38:25 Join midgey [0] (
23:38:56greatgatsbyquick question: i used ubuntu gutsy to transfer songs to 30G U2 ipod video, and the transferred songs played with ~<2 second inserts of other songs in the ipod
23:39:34 Quit ender` (Read error: 110 (Connection timed out))
23:40:10rashergreatgatsby: I've seen this, and I think it's a Linux problem (happened with my Iriver H120, which uses a hardware usb bridge)
23:40:21saratogai believe i have worked all bugs out of the default theme patch (FS #8443)
23:40:27rasherOr at east Linux-related.
23:40:33saratogai'd appreaciate it though if someone could look over it before i submit
23:40:43saratogaalso, testing on a target with remote would be nice, since i don't have one of those
23:41:13Lloreangreatgatsby: You should find the songs *on the ipod* and try playing those files back on your PC
23:41:36amiconngevaerts: Neither. HAVE_FLASH_STORAGE doesn't define which driver is used, it defined behaviour
23:41:37magmaniacrasher: funny, I've been using a H120 with Linux for years, and never encountered such behaviour
23:42:08amiconnE.g. the ipod nano does define HAVE_FLASH_STORAGE, while it does use the ata driver
23:42:53greatgatsbysongs played with inserts when transfered from the ipod...but original files are ok
23:43:21rashermagmaniac: you might have to do something stupid to trigger it, or perhaps certain hardware triggers it
23:43:28magmaniacgreatgatsby: this sound like the filesystem on the ipod is damaged.
23:43:30linuxstbgreatgatsby: Sounds like a filesystem corruption - do you always unmount the ipod before unplugging it?
23:43:43greatgatsbyyes..i always unmount
23:44:22greatgatsbyooops..that sounded funny. lol
23:44:22 Quit saratoga ("CGI:IRC (EOF)")
23:46:44 Join csc` [0] (n=csc@archlinux/user/csc)
23:46:52magmaniacgreatgatsby: can you do a filesystem check of your ipod?
23:47:34greatgatsbyi will
23:47:34gevaertsamiconn: ok. Can I assume that ata_get_identify() will be available on non-ipod, non-sansa usb stack using targets ?
23:49:26amiconnRight now you can, I think
23:50:39 Quit tvelocity (Remote closed the connection)
23:52:51 Quit jgarvey ("Leaving")
23:54:53 Quit MajorC ()
23:58:23 Join crzyboyster [0] (n=4b596e0c@gateway/web/cgi-irc/
23:58:31 Join Lars_G [0] (n=Lars@unaffiliated/lars-g/x-000001)
23:58:49Lars_GHey all, what's the "rigth" place to set USE_ROCKBOX_USB, autoconf.h on my build dir?

Previous day | Next day