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 2008-07-07

00:00:29 Quit hannesd (No route to host)
00:00:40 Quit Zagor ("Client exiting")
00:01:09 Quit vort3x (Read error: 104 (Connection reset by peer))
00:02:06 Join CyBergRind|w [0] (n=cbr@
00:07:48mcuelenaereis memmove necessary for Rockbox to work?
00:09:58stripwaxmcuelenaere - as in, does Rockbox currently use it, or could Rockbox have been implemented without the need for it?
00:10:31mcuelenaereas in: I'm trying to get a MIPS target to run in Rockbox, will it need it?
00:11:01mcuelenaereit seems most use for memcpy and memmove the same function on MIPS targets
00:11:41stripwaxmemmove needs to be able to handle overlapping source and target, but if the mips implementation of memcpy handles that already then you can alias memmove to it
00:14:19 Quit mf0102 ("Ex-Chat")
00:15:02 Join jhulst__ [0] (
00:17:03 Join vort3x [0] (n=vortex@unaffiliated/dfa001)
00:19:25 Quit [CBR]Unspoken|w (Connection timed out)
00:19:27 Quit davina ("GNU/Linux the free alternative to Windows")
00:20:41 Quit shotofadds (Read error: 104 (Connection reset by peer))
00:23:28 Quit Lear ("ChatZilla 0.9.83 [Firefox 3.0/2008052906]")
00:23:58 Join Preston [0] (
00:29:29 Quit bluebrother ("leaving")
00:29:32 Quit Vorador ("Ex-Chat")
00:32:13 Quit moos (Read error: 110 (Connection timed out))
00:36:21 Quit lard-Vader (Read error: 110 (Connection timed out))
00:36:29 Join lard-Vader [0] (
00:36:38Prestonits like potato chips and masturbation
00:37:28stripwaxPreston - this doesn't sound like the start of an on-topic conversation...
00:37:45Prestonmy bad
00:37:45 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
00:38:17Prestonbut ya just cant stop
00:38:45stripwaxLear/preglow - enabling the same iram defines for pp502x as used by coldfire give only a very small speedup - about 0.5 MHz
00:39:19 Part Preston
00:39:48 Quit herrwaldo (Remote closed the connection)
00:40:35stripwaxLess on average, in fact
00:43:45***Saving seen data "./dancer.seen"
00:45:43 Quit ender` (" "Religion is the opiate of the masses." -- Karl Marx )( "Winners don't do drugs." -- The FBI")
00:46:20 Join BilleniumLAPTOP [0] (
00:46:44BilleniumLAPTOPCan someone link me to the mpeg plugin wki?
00:47:04*mcuelenaere has full LCD access on his VX747 :)
00:48:40dionoeaBilleniumLAPTOP: can't you use the search box in the wiki?
00:48:46BilleniumLAPTOPi have searched
00:48:50 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
00:48:58stripwaxsearch for 'mpegplayer'
00:49:03dionoeahum, PluginMPEGPlayer should be the page name if i remember correctly
00:49:23 Quit ompaul (Client Quit)
00:49:25dionoeaand congrats to mcuelenaere even though I don't know what that player is :)
00:49:52*stripwax wonders *which bit* of III_decode is slow...
00:50:32 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
00:50:42stripwaxevidently none of the subroutines, unless they've been inlined ..
00:51:24dionoeaTouchscreen \o/
00:51:40 Quit miepchen^schlaf ()
00:51:43*dionoea 's currently writing a touchscreen paint program for TomTom Go GPS
00:52:01stripwaxSomeone should port xmerlin for on-screen keyboard ;-)
00:52:15 Join miepchen^schlaf [0] (
00:52:39dionoeahehe, that player looks like an iPod Touch ripoff :)
00:53:10mcuelenaereindeed :) it is Chinese ;)
00:53:56 Part pixelma
00:54:48 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
00:56:15saratogastripwax: hadn't realized that changing the lowpass filter in lame actually changed what synth_full did
00:56:18 Quit robin0800 (" HydraIRC -> <- Go on, try it!")
00:56:32stripwaxyep, looks like it
00:56:43saratogapreglow: the 64 bit multiplies in synth_full aren't a huge deal
00:57:26saratogathey're actually no slower then doing 32 bit multiplies there because they save you 2 cycle for renormalizing back to 4.28 fixed format
00:57:39saratogai think its slow mostly because synth_full does:
00:57:49saratoga1) a whole lot of multiplies, and multiplies are really, really slow
00:58:38saratoga2) the data format seems to involve a lot of random memory reads, so it can't make very effective use of ldm instructions, which means a lot of the loads take 2x as long as they could
00:59:07 Quit jhulst__ (Read error: 110 (Connection timed out))
01:08:05 Quit Nico_P (Remote closed the connection)
01:12:15 Quit PaulJam (".")
01:13:53 Join bughunter2 [0] (n=Jelle@
01:27:04 Quit jhulst_ (Remote closed the connection)
01:28:12 Join delorean90 [0] (
01:33:11delorean90may i have write permisson to the rockbox wiki?
01:37:23*stripwax has just noticed that asm routines don't seem to get instrumented for profiling - should've realised earlier
01:37:52mcuelenaeredelorean90: what's your Wiki name?
01:41:14 Join Thundercloud [0] (
01:41:24 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
01:41:49 Quit gregzx ("ChatZilla 0.9.83 [Firefox 3.0/2008052906]")
01:43:03 Quit BilleniumLAPTOP ()
01:44:16*delorean90 just realized he forgot to enable flashing for this channel
01:44:52mcuelenaeredelorean90: what do you want to edit?
01:45:19delorean90i would like to be able to help keep the sansa view wiki in order as things progress
01:46:26mcuelenaereYou're added! Promise not to spam :)
01:46:57delorean90thank you very much, and i would never spam, i like being a member of rockbox too much :)
01:46:58mrkikoand not ruin the work of others
01:48:10delorean90id probably be so nerveous of that that i would show what i was writing and make sure its ok he he
01:48:15 Quit tvelocity ("Αποχώρησε")
01:52:32mrkikodelorean90<: I don't partecipate to avoid this :)
01:52:39 Quit culture (Read error: 110 (Connection timed out))
01:52:54mrkikobut I think it's good doing so - helping in developing rockbox is good
01:53:24delorean90ill mostly only edit if im asked to like " i just finsihed geta this datasheet could somebody add it to the wiki"
01:57:47 Quit mcuelenaere ()
01:58:16 Join kushal_12_27_200 [0] (
01:58:58 Join spiorf [0] (
01:59:09 Quit spiorf (Remote closed the connection)
02:00:45 Join spiorf [0] (
02:01:22 Join Shaid [0] (
02:02:54 Quit spiorf (Read error: 104 (Connection reset by peer))
02:03:02 Join xqtftqx [0] (i=4cbd373c@gateway/web/ajax/
02:03:12 Join spiorf [0] (
02:03:53xqtftqxWhould anybody be intrested in making something for rockbox wich allows icons in indivual files insted of all lined up?
02:05:33kushal_12_27_200what would I have to do?
02:06:00xqtftqxidk, just help out, some programing experience whould be needed
02:06:30delorean90what do you mean by incons in individual files?
02:06:35 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
02:06:49xqtftqxthe menu icons
02:07:05xqtftqxthey whould have there own files, like "files.bmp database.bmp etc
02:07:14delorean90well yes, but arent there icons already
02:07:23delorean90ohh ok
02:07:25xqtftqxyeah but in 1 file
02:07:42xqtftqxmaking it hard for somebody to make custom icons for a theme
02:10:56xqtftqxanybody know if theres a psd avaible for the cabbie theme?
02:12:02 Quit vort3x (Read error: 104 (Connection reset by peer))
02:12:30 Join vort3x [0] (n=vortex@unaffiliated/dfa001)
02:18:58 Quit Shaid ("*toink*")
02:28:48 Join Thundercloud_ [0] (
02:29:31 Quit nuonguy (Connection timed out)
02:39:07 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
02:39:30 Quit xqtftqx (" ajax IRC Client")
02:39:51 Join vincenzo [0] (
02:43:21vincenzoHi, I'd like to use rockbox... where can I buy a compatible player?
02:43:46***Saving seen data "./dancer.seen"
02:44:03delorean90one second ill give you the wiki link to the buyers guide
02:44:23 Quit MethoS-- (Remote closed the connection)
02:44:30delorean90 hope this helps :)
02:45:21vincenzoI'd like the sandisk ones, but V2 is unsupported. And the others are no more in production :-(
02:46:36delorean90sometimes you can find them on ebay and other random stores. the ipod video is being sold on apples site for 179
02:47:29delorean90my advice if you really fancy a sansa, is to buy one of thier new ones that already has a "new port" thread in the rockbox forums, and just follow it to see if rockbox gets ported
02:47:38 Quit Thundercloud (Read error: 110 (Connection timed out))
02:48:31vincenzowould you personally advice me buying something else?
02:49:07 Quit CaptainKewl (Remote closed the connection)
02:49:40delorean90well what do you mean? if you really want rockbox right away, i suggest the video or really any other player supported that you can find. otherwise do some searching around the new ports section of the forums and see which one has the most progress
02:52:56delorean90 if you take a look at the new ports section those are some of the daps that will almost definitely be ported *eventually*
02:54:19 Quit HellDragon (Client Quit)
02:55:12kushal_12_27_200you guys at rockbox are great. I am very grateful to you
02:55:23 Quit kushal_12_27_200 ("Leaving")
02:55:34 Join nuonguy [0] (
03:05:26lard-VaderIn case no-one has pointed this out, the boxes you make rock.
03:08:29 Join HellDragon [0] (
03:12:17 Quit bughunter2 ("bye")
03:13:19lucentlard-Vader: I get enough about that in my email spam thanks
03:13:54 Join xqtftqx [0] (i=4cbd373c@gateway/web/ajax/
03:15:32vincenzodelorean90: I'd like a player that is "good quality", quite small, and that can run rockbox... I'd like one that is impossible (virtually) to brick...
03:16:04xqtftqxhas anybody used winff in linux to convert to rockbox? i get a error "unsported codec..."
03:16:27xqtftqxvincenzo: I whould recomend a gigabeat, that you can find on ebay, or a sansa
03:17:03vincenzoxqtftqx: the sansa (e280v1) is what I would like...
03:17:15vincenzoxqtftqx: How do I make sure I byu a V1?
03:17:41xqtftqxlook for "V1" somebody on the box or ebay auction
03:17:50xqtftqxor contact who ever is selling it
03:18:13MarcGuayvincenzo: The only way to know for sure is the firmware version. You'll have to find one second hand and ask the current owner.
03:19:32delorean90vincenzo, and as long as you always follow instructions, its almost impossible to brinck your supported dap
03:20:59vincenzoShould I chose one from the "new ports" which one is in better shape in your opinion?
03:21:23xqtftqx"new ports" means there in devolpment
03:21:35xqtftqxrockbox runs but isnt ready for the general public
03:21:56MarcGuayvincenzo: The Gigabeat S is making good progress... The D2 as well, but still a lot of work before it's functional... I wouldn't recommend any of them if you want a sure thing.
03:22:20delorean90only chose a dap from that list if your prepared to wait a bit for support
03:22:43xqtftqxGigabeat S is going realy nice, i think all we need is a manual and bug fixes
03:22:52MarcGuayvincenzo: And even then, there's no guarantee they're even be finished.
03:23:21xqtftqxits a 99.9% chance they will be
03:23:27*MarcGuay should quit while he's ahead.
03:23:28xqtftqxah lets say 98
03:23:45 Quit lard-Vader ("Leaving.")
03:24:05MarcGuayvincenzo: If you buy one and it's never finished, remember this name: xqtftqx (very memorable). :)
03:24:27*delorean90 hopes for a port to the view
03:24:47vincenzoI can understand that... as a (very minor) KDE contributor I know how OSS developement goes :-)
03:25:32xqtftqxGrrr, i cant convert a video in linux!
03:26:19MarcGuayvincenzo: If you have access to eBay you can still find Gigabeat Fs, e200s, c200s, iriver h100/h300s... iAudio's have gotten very pricey but show up now and again...
03:26:19xqtftqxah, ill use windows :(
03:26:25vincenzoBut if rockbox does somehow work (and not brick it) I could always get involved in improving (2% of chance I have enaught time)
03:27:24xqtftqxvincenzo: No worrys in bricking
03:27:47vincenzoI think I'll eventually go for the e280vx...
03:27:54xqtftqxGood Choice
03:28:01*delorean90 thinks back to when the iriverh10 was first ported and smiles
03:28:46vincenzomaybe I'll try on pixmania or ebay. If I get a v2 (I'll delay the rockbox adoption:-(((( )
03:29:19vincenzoDoes v2 read SDHC at least?
03:29:26vincenzoI fear not !
03:29:50vincenzos/I fear not !/I fear it won't/
03:30:24advcomp2019vincenzo, yea the v2 can read sdhc should but this was getting off topic
03:31:05vincenzoadvcomp2019: :-) fair enaught :-)
03:31:43vincenzoHow many people are involved (if you know them) in v2 porting?
03:32:04vincenzoI'm just reading trught the forum.
03:32:09delorean90vincenzo, if you would like to discuss more about daps feel free to join us on the #rockbox-community channel
03:33:06saratogai think theres a couple people interestd in the V2, but there isn't really any porting yet
03:35:13 Quit DerDome (Nick collision from services.)
03:35:14 Join DerDome1 [0] (
03:35:26 Nick DerDome1 is now known as DerDome (
03:48:40 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
03:51:01lucentxqtftqx: the commandline arguments to convert video using ffmpeg or mencoder are something of a magical artform, the way I look at it
03:51:45lucenteach (interface) links to many different video codec libraries, with their own arguments
03:51:52lucentand no help feedback system
03:52:24lucentthere is never -avcfopts audio=help
03:52:41lucentinstead the man page says "see xyz project source code for details"
04:04:21MarcGuayrasher, or someone else with the knowhow: Could you refresh the wiki page or teach me how to do it myself?
04:05:01 Quit shadoxx (Remote closed the connection)
04:08:06 Quit saratoga ("CGI:IRC (EOF)")
04:13:24 Quit axionix (Read error: 104 (Connection reset by peer))
04:14:28 Join Thundercloud [0] (
04:22:46 Join axionix [0] (
04:24:41 Quit fyrestorm (Read error: 104 (Connection reset by peer))
04:24:59 Join fyrestorm [0] (
04:25:07 Quit jhulst (Read error: 113 (No route to host))
04:25:51 Join fyre^OS [0] (
04:26:28 Quit agaffney (Read error: 104 (Connection reset by peer))
04:27:13 Join ivor [0] (
04:28:16 Quit Thundercloud_ (Read error: 110 (Connection timed out))
04:28:27MarcGuayAnyone with an iPod around?
04:29:07 Quit vincenzo (Remote closed the connection)
04:29:16ivorWould someone please grant me wiki write permissions? wikiname is JasonTerhune
04:29:53MarcGuayivor: Mind if I ask what you plan to change? Just curious...
04:29:55 Join J-23 [0] (
04:30:20ivorsure Marc, I have a WPS update to upload
04:30:58MarcGuayivor: Alrighty. Is the copyrtight license safe & sound?
04:31:50ivoryeah, I updated an existing WPS
04:31:54ivorcreative commons, share alike
04:32:07ivorIt's the Ultima theme
04:32:10MarcGuayivor: Excellent. You're ready to roll...
04:32:15ivorGreat, thanks!
04:35:52 Join SirFunk_ [0] (
04:36:44MarcGuayI'm looking for people with ipods (any model) or Archos players to test the A-B repeat keypresses, specifically the keys to set the A/B points and reset the markers.
04:36:54 Join agaffney [0] (n=agaffney@gentoo/developer/
04:38:09 Join AndyIL [0] (n=pasha_in@
04:38:28 Quit Thundercloud (Remote closed the connection)
04:41:36 Join axionix_ [0] (
04:42:04 Quit axionix_ (Client Quit)
04:42:21 Join axionix_ [0] (
04:42:23 Quit axionix_ (Client Quit)
04:43:49***Saving seen data "./dancer.seen"
04:43:54 Quit fyrestorm (Read error: 113 (No route to host))
04:44:11 Join fyrestorm [0] (
04:45:59 Quit fyre^OS (Read error: 113 (No route to host))
04:49:31 Quit AndyI (Read error: 110 (Connection timed out))
04:51:17 Join miepchen^schlaf_ [0] (
04:54:13 Quit SirFunk (Connection timed out)
04:57:22 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
04:57:35 Join agaffney_ [0] (n=agaffney@gentoo/developer/
04:57:42 Quit agaffney (Read error: 104 (Connection reset by peer))
04:57:44 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:59:06 Join Shaid [0] (
05:02:18 Join cool_walking_ [0] (
05:07:26 Join fyre^OS [0] (
05:07:26MarcGuayContinuing the request above, information for the X5/M5 would also be helpful.
05:10:50 Join JdGordon|uni [0] (i=768ae017@gateway/web/ajax/
05:11:30JdGordon|uniMarcGuay: hey, you could use the sims to check them... they are "close enough"
05:15:33 Nick agaffney_ is now known as agaffney (n=agaffney@gentoo/developer/
05:16:24 Part J-23
05:24:45 Quit fyrestorm (Read error: 113 (No route to host))
05:24:58 Join fyrestorm [0] (
05:26:25 Join J-23 [0] (
05:28:53 Quit fyre^OS (Read error: 113 (No route to host))
05:30:03 Part J-23
05:35:34 Quit JdGordon|uni (" ajax IRC Client")
05:35:55MarcGuayJdGordon: Thanks man, I always forget about them...
05:41:46MarcGuayJdGordon: Hmm. I can't seem to find the place in gwps.c where targets without a ACTION_WPS_ABSETB_NEXTDIR call the ab_set_A/B_marker. On the Nano select does this but I can't find it anywhere...
05:50:23 Quit homielowe (Remote closed the connection)
05:51:34 Join homielowe [0] (
05:53:08MarcGuayIt would be helpful if these things were all in the same place, though I understand that targets are different and code is hard to maintain. It's frustrating for people trying to get into it to get roadblocked like this. Not pointing any fingers as I wouldn't have a clue where to point em, just sayin' is all.
05:58:49 Join J-23 [0] (
05:59:11 Join sarixe [0] (
05:59:32 Quit DataGhost (Nick collision from services.)
05:59:40 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
06:05:34 Join shadoxx [0] (
06:10:27 Quit Schmogel (Read error: 110 (Connection timed out))
06:11:55JdGordonMarcGuay: yep.. thats definatly a problem :p
06:12:07 Quit delorean90 ()
06:13:10 Quit Zarggg ()
06:13:16JdGordonMarcGuay: hmm... your recent commit seems to have unrelated changes to the message?
06:13:23LloreanIsn't that more because a lot of the cold is just old, and hasn't been adapted to being more centralized?
06:14:43JdGordonyeah, no.. i was agreeing with MarcGuay there.. not saying that the code has bugs...
06:15:15 Part J-23
06:15:49 Join x30 [0] (n=d2dc468e@gateway/web/cgi-irc/
06:16:27JdGordonbon jour!
06:16:35JdGordonohayo gozaiymas
06:16:40 Quit HellDragon (Read error: 104 (Connection reset by peer))
06:16:46JdGordonasalam alekem
06:16:56 Join HellDragon [0] (
06:17:04x30ah.. I can speak english
06:17:41x30whatever... is there anyone who knows what the rescue mode is?
06:17:58 Join Preston [0] (
06:18:06x30my gigabeat X30 with rockbox has freezed since I put the gbsystem folder on it
06:19:20x30in addition it isn't recognized on PC when I connect gigabeat with usb
06:19:20lucentx30: hi, I don't know. the Gigabeat dev goes by "xqtftqx" I think
06:19:33 Part Preston
06:20:10x30hmm.. then do you know what the rescue mode is? I don't know exactly
06:20:30lucentno, actually I don't have a player capable of rockbox
06:20:44lucentI can't try it out like I would want to and give you an answer from experience
06:21:16x30ah.. ok thanks whatever :)
06:21:35 Quit x30 ("CGI:IRC (EOF)")
06:21:40Lloreanlucent: xqtftqx isn't a dev...
06:21:48lucentagh, my mistake
06:21:58lucentLlorean: what was the correct answer?
06:22:26LloreanThere aren't really player-specific devs.
06:22:38LloreanMost devs have done work on many specific players.
06:22:57LloreanGenerally only early on is there a dev or a few devs specific to any given player, but that's just when the port is very new.
06:23:41lucentI'm still searching for a hardware device to run rockbox on
06:23:45LloreanAs for the Gigabeat X, it's only a nominally supported target. It _should_ work, but we don't actually have any regular dev who owns one, so nobody can really perform regular testing, or tweaks for its few differences from the Gigabeat F
06:23:56Lloreanlucent: Seen the BuyersGuide page in the wiki?
06:24:25lucentit sounds kind of risky though, like v1 / v2
06:24:55lucentthere's nothing in production
06:26:08LloreanThe v1/v2 issue is only with the Sandisk players we support
06:26:13LloreanEverything else is pretty explicit.
06:26:44lucentwell, the iPods too. which Gen, not all sellers know this 2nd hand
06:28:28lucentsansa e200v1 looks nice, wish I could find one for sale and know it would be a v1 :)
06:28:39LloreanIt's really, really easy to tell the difference between iPod gens.
06:33:56MarcGuayJdGordon: Yeah I goofed on the commit with unrelated changes and got some tips on how to avoid it in the future.
06:34:30JdGordonok :)
06:34:55MarcGuayAnyone with thoughts on where the ipod/x5/m5 A-B repeat code is hiding?
06:36:23LloreanHave you tried to grep for it?
06:37:17MarcGuayYeahm grep is my new best friend. No dice, at least for the WPS_ABRESET button.. I should try with the actual function, though...
06:37:35LloreanProbably a better bet
06:38:49MarcGuayNo, I tried that too. It only exists in gwps.c and abrepeat.c/h
06:39:47MarcGuayI have a feeling it's coded in a different way. They use a 1st select = set A, 2nd select = set b, 3rd select = reset, unlike the combos for other targets.
06:40:05 Quit miepchen^schlaf_ ()
06:40:17MarcGuayI suppose I could just throw it in the manual after testing on the sims but now this is bugging me...
06:40:26 Join _vertic23 [0] (
06:42:15LloreanMarcGuay: You could just check the "since 2.5" page, and look for mention of "repeat" or "a-b repeat" on it, and look for one that either touches other files, or also mentions iPod
06:42:16MarcGuay... And I couldn't find any info for the Archos player..
06:42:44lucentwow, off-topic... how does a company end up with a $10,000usd audio player?^~^HEWLETT-PACKARD
06:42:44MarcGuayLlorean: Thanks for the tip.
06:43:03lucentmust be a database problem
06:43:10LloreanMarcGuay: It's not the most efficient thing ever, but it might work.
06:43:22JdGordonMarcGuay: ok, not busy anymore.. what you looking for?
06:43:38Lloreanlucent: Please, stay on-topic in here. There's a channel specifically for off-topic stuff. This one's logged, and we like to keep them focused.
06:43:40MarcGuayJdGordon: Read back a bit.
06:43:51***Saving seen data "./dancer.seen"
06:44:07JdGordondo those targets have ACTION_WPSAB_SINGLE defined?
06:44:15lucentLlorean: thank you, I'll make myself scarce. :)
06:44:21 Part lucent
06:45:22MarcGuayJdGordon: I can't find it in keymap-ipod.c
06:45:27 Quit vertic23 (Read error: 104 (Connection reset by peer))
06:46:11JdGordonMarcGuay: ok, yes they do... its in the config files for ipod, miezu m6, x5/m5/m3.... in a-b mode the first press of the browse button sets A, then B, then resets...
06:46:19JdGordonlines 241-261 in gwps.c
06:47:00JdGordonno mention of the archos players though
06:47:35MarcGuayJdGordon: Arg. No wonder it was so hard to find, it resets the button instead of calling the function by name... Thanks hombre...
06:48:35*MarcGuay ponders a way to bring those things closer together..
06:48:59JdGordonna, its fine as it is
06:50:07MarcGuayYou think? It's a mess to me.
06:50:41JdGordonoh I agreee its a mess.... I just think its fine being a mess....
06:50:45MarcGuayGranted, I can't think of another way to do it.. :)
06:51:15MarcGuayYou can't define the button twice for the same context so it has to be done elsewhere.
06:52:15MarcGuaySo any ideas on how the player sets the A/B points?
06:52:32JdGordondoes the player even have a-b mode?
06:52:59MarcGuayJdGordon: It does on the sim.
06:53:44JdGordon" You can't define the button twice for the same context so it has to be done elsewhere." ??
06:54:43MarcGuayJdGordon: Or maybe you can? Could ACTION_WPSAB_SINGLE be moved from the configs to the keymaps where it makes more sense?
06:55:28JdGordonoh, woops, now I get you... yes, the same button cant return 2 different actions
06:55:28MarcGuayI think I'm confused, ignore me.
06:56:24JdGordonand yep, I cant see how to set a/b points on the player either
06:56:45MarcGuaySo the player remains a mystery. Hopefully someone will read this in the logs and send out a bat-signal.
06:56:56JdGordondont expted it...
06:57:46MarcGuayWash, Rinse, Repeat. Gotcha.
07:15:46 Join SirFunk__ [0] (
07:18:37 Quit SirFunk_ (Read error: 110 (Connection timed out))
07:33:42mrkikoHi all!
07:37:49 Part lolmaus_Melven
07:38:14JdGordonMarcGuay: lol
07:38:28JdGordonalso, that note is in action.h also...
07:41:41 Join TemptationToast [0] (
07:43:23 Part TemptationToast
07:49:20 Quit MarcGuay ("ChatZilla 0.9.83 [Firefox]")
07:57:32amiconnJdGordon: Hmm, looks like the player does have ab repeat enabled, but it's not usable because it's not mapped
07:58:03*amiconn wonders how that single mode works on iaudios and ipods
07:58:08JdGordonyep, we figured that out...
07:58:12*amiconn never used ab repeat so far
07:58:22*JdGordon neither
07:58:26JdGordonother than testing
07:58:48amiconnab repeat buttons for the player could be mapped (as combos) btw - I just don't know yet how to map the reset
07:58:50JdGordondoes ondio still not have a-b also because there is no buttons for it?
07:59:08amiconnBut it could also use the single mode if that works
07:59:21JdGordonit should work
07:59:28JdGordonhavnt heard any complaints about it
07:59:33amiconnOn Player, ON+LEFT and ON+RIGHT aren't taken yet in the wps
07:59:54JdGordonplayer doesnt have dirskip also?
07:59:54amiconnBut single mode may be better, as cobs are nasty one-handed
08:00:29amiconnUmm, I don't know
08:00:37amiconnAnother feature I never use
08:00:56amiconnBut I think it should, as it doesn't need extra button mappings
08:07:10 Join gevaerts_ [0] (
08:10:53 Join Buschel [0] (
08:13:28Buschelstripwax/preglow/saratoga: (for the logs) the synth (dct32+synthfull) might be faster on mp3-96kbps because of several subband being zero'ed. through this you'll have more muls with 0 at least in the dct and smaller coefficients (less energy) in synthfull.
08:14:53Buschelthe same should happen in the entropy coding and requantization. additionally there should be lots of intesitiy stereo coding which saves a channel in the upper bands −− at least in the entropy cding
08:16:01Buschelbtw, some days ago I've merged the libmad dct into mpc. compiling the C-version of dct32 is a bit faster than the asm-version on PP5022
08:16:51BuschelC-version compiled with -O1 is about 0.8MHz faster than the asm'ed version
08:18:07 Quit gevaerts (Read error: 110 (Connection timed out))
08:18:26Buschelthe libmad c-dct is faster than mpc's dct, but mpc's dct is faster than libmad's asm-dct
08:19:42Buscheli'll try to make a fine and proper merge (without any additionally interfacing code like resorting afterwards, etc.) of libmad's c-dct into mpc.
08:20:14amiconnThe asm dct for libmad was clearly faster than the C version when it was added
08:20:16Buschellibmad's c-dct has a big advantage over mpc's one -> it is far less sensitive to internal overflows
08:21:48Buschelamiconn: from what i've read (as far as i remember in the fs-entry) the asm'ed version was claimed to faster on some processors because it was much smaller code, but not optimized for performance in terms of operations
08:22:05Buschel*claimed to be...
08:22:46 Join [1]cool_walking_ [0] (
08:24:32amiconnBuschel: tomal wrote it, in order to get libmad realtime on iFP (which is only 60MHz ARM7TDMI)
08:24:57Buschelamiconn: check out fs#6705
08:25:34Buschel"smaller in iram but worse in terms of executed instructions"
08:26:12Buschelat least I should check it with -O1 on my PP
08:28:08amiconnAha, but the tests all show that the asm version is faster
08:28:08 Quit homielowe (Read error: 104 (Connection reset by peer))
08:28:54 Join homielowe [0] (
08:31:28Buschelamiconn: the patch changed synh_full + dct32. i can see no test result for the dct32-part only...
08:31:34*Buschel gotta go to work now
08:31:39 Quit Buschel ()
08:36:58 Join Rob2223 [0] (
08:37:39 Join stoffel [0] (
08:39:20 Quit cool_walking_ (Read error: 113 (No route to host))
08:39:20 Nick [1]cool_walking_ is now known as cool_walking_ (
08:43:52***Saving seen data "./dancer.seen"
08:46:00 Quit m0f0x ()
08:48:51amiconnJdGordon: I just tried AB repeat in single button mode on the Player. It's working nicely (but you don't get feedback for the first press... something that can be fixed later)
08:49:07amiconnI'll try AB repeat on Ondio in single button mode too, then commit
08:49:39JdGordonok, cool
08:49:48JdGordonfix the manual while your at it.. :D
08:52:31 Join BigBambi [0] (i=86ceaf40@gateway/web/ajax/
08:54:58 Quit Rob2222 (Read error: 110 (Connection timed out))
08:56:08 Quit tessarakt ("Client exiting")
09:08:45 Quit perrikwp (" ajax IRC Client")
09:10:19 Join pixelma [0] (i=d5494c39@gateway/web/ajax/
09:12:35pixelmaamiconn, JdGordon: I remember the AB-repeat was left disabled on Ondios because it added like 1k something bytes to the binary and broke rombox at that time. Also I think I remember that resetting the markes didn't work because it was mapped to a double press of mode but the single press is already mapped (return to file browser)
09:12:46pixelmarombox on OndioFM that was
09:13:51 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
09:13:52JdGordonok, well its up to you if you want it reenabled... I dont tihnk anyone else uses ondio :p
09:14:18 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
09:14:53JdGordonhey B4gder, any idea why isnt being used? I did a reinstall a few days ago but have the rbclient setup again and it should be workling
09:15:37B4gderI'll check
09:16:10B4gderah new key
09:16:49*GodEater thinks should be fine to work as a server again now too
09:17:21JdGordonah, thought that might be the case... where can i get the new key from?
09:17:37GodEaterJdGordon: it's *your* new key that's the problem I suspect
09:17:43GodEaternot one you need from B4gder
09:17:51B4gderfixed now
09:18:14*JdGordon is a dill
09:18:17JdGordonok, cool
09:21:19amiconnpixelma: No double presses involved. AB repeat in single button mode just cycles through the operations (set A, set B, reset markers)
09:21:54amiconnWhen enabled, it removes the ability to directly enter the browser, but that's a tradeoff. The browser is still reachable via the menu
09:22:10amiconnThat applies to all targets with AB repeat in single button mode, not just Ondio
09:24:16pixelmahmm, then I remembered wrong
09:24:51 Join petur [50] (n=petur@rockbox/developer/petur)
09:26:43 Quit B4gder (Excess Flood)
09:27:18 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
09:27:35JdGordonamiconn: the feeling there is that if your in a-b mode you wouldnt need the browser...
09:27:36 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
09:30:38 Quit pixelma (" bbl")
09:42:51 Quit jon-kha (Read error: 145 (Connection timed out))
09:45:07 Join safetydan [0] (n=safetyda@rockbox/developer/safetydan)
09:47:26 Quit Shaid (Read error: 60 (Operation timed out))
09:54:26 Quit shadoxx (Remote closed the connection)
09:56:04safetydanhrm, is a single line fix worth an addition to the credits file?
09:56:47 Join K4rP4D [0] (n=KrPD@unaffiliated/krpd)
09:56:47B4gderyes imho, it's easy to be generous with those things
09:57:59 Quit K4rP4D (Client Quit)
09:58:29 Join K4rP4D [0] (n=KrPD@unaffiliated/krpd)
09:59:42 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
10:00:23safetydanB4gder, I agree. Especially since there's investigation time as well for finding the fix.
10:00:46B4gderyes, very true
10:02:13*GodEater seems to recall he got into CREDITS based on about four lines of rather trivial code added to ipodpatcher
10:06:02 Join Nico_P [0] (i=53915df2@gateway/web/ajax/
10:08:50*B4gder recalls being the one who wrote and added the first CREDITS list ;)
10:09:42GodEaterthat's definitely a trivial way to get into it ;)
10:10:27B4gderyeah, rather lame I agree ;-)
10:11:20 Join ender` [0] (
10:19:53JdGordonstill no :(
10:21:07B4gder"svn: Working copy '.' locked"
10:21:09B4gderme fix
10:21:31JdGordonhehe.. i heard the disk clicking away and thought you started another build
10:21:46*JdGordon wonders why it was locked?
10:22:09B4gderthat happens at times when it gets interrupted etc
10:22:48peturI used to have that with cygwin, never seen it under linux
10:23:38B4gderI've also seen it happen when you try to svn up the same repo from multiple processes...
10:24:41 Part homielowe
10:27:20markunB4gder: the git repo works really nice, but do you know a nice way for me to commit something?
10:28:15markunI hope that after a "git pull" I wouldn't have to do a "git svn rebase"
10:29:02B4gderI don't know much git stuff, I just made the mirror thing work with the help of others' instructions
10:29:26Nico_PB4gder: have you found out how to make it accessible via HTTP?
10:30:19B4gderno, I was away during the weekend and was dead tired yday night so I haven't really attacked that yet
10:32:04Nico_Pno rush anyway
10:36:37 Join Genre9mp3 [0] (n=yngwiejo@rockbox/contributor/Genre9mp3)
10:40:32 Join desowin [0] (n=desowin@atheme/member/desowin)
10:43:09 Join davina [0] (
10:43:53***Saving seen data "./dancer.seen"
10:44:24 Join UncleRem1s [0] (
10:50:23 Nick gevaerts_ is now known as gevaerts (
10:52:35 Quit GodEater (" ajax IRC Client")
10:54:07 Quit UncleRemus (Read error: 110 (Connection timed out))
10:59:21 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
11:01:06 Join GodEater [0] (i=c2cbc962@gateway/web/ajax/
11:01:06 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
11:08:57 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
11:14:50 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
11:18:12 Quit cool_walking_ (" ")
11:20:52 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
11:22:49B4gdermcuelenaere: ah right, the thread stuff in the kernel needs some mips attention
11:23:46 Quit spiorf (Remote closed the connection)
11:24:30mcuelenaereB4gder: indeed, and it isn't that easy as I'm not really familiar with MIPS
11:25:09 Join bertrik [0] (n=5a80bda4@gateway/web/cgi-irc/
11:25:17B4gderI can't say that I am either, even though I actally own a mips book and have poked a bit on mips before...
11:25:52mcuelenaerewell at least you can verify my code :)
11:27:34mrkikoThe r17951 release crashe already three times since I installed it :)
11:28:14mrkikobut the third time, pressing play the playter rebooted
11:28:28 Quit mcuelenaere (Read error: 131 (Connection reset by peer))
11:28:49 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
11:29:09 Join PaulJam [0] (
11:29:40mcuelenaereis there a way to build a bootloader build with threading disabled?
11:30:27B4gderI don't know
11:30:46mcuelenaereok, then I'll just comment it out in SOURCES
11:37:23mcuelenaeredoes Rockbox normally need includes from not within itself i.e. /usr/lib/gcc/mipsel-linux-gnu/4.2.4/include/.. ?
11:39:04*mrkiko thinks not
11:39:54mcuelenaeredoes a compiler set include paths itself and if so, how do I unset them?
11:42:46 Quit mcuelenaere (Read error: 54 (Connection reset by peer))
11:43:00*JdGordon wonders if he remembers any mips from first year uni...
11:43:17 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
11:45:57 Quit bertrik ("CGI:IRC (EOF)")
11:52:24 Join Lear [0] (i=chatzill@rockbox/developer/lear)
11:52:51*mcuelenaere wonders why GCC has a limits.h in it's include path which includes <limits.h> but can't find it..
11:53:17mrkikoit seems strange
11:54:13mcuelenaereprobably whenever a file is included, it can't be included again in the same gcc process
11:54:38mcuelenaerebut apparantly this file does want to include a file with the same name as itself
11:55:37mrkikobut distinctions are made between paths
11:55:53mrkikoa/something.h can include b/something.h
11:56:03mrkikoeven if they are something.h
11:56:11gevaertsa file can include itself if it wants to
11:56:43 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
11:56:50 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
11:57:15 Quit stoffel ("leaving")
11:57:27mcuelenaere and are what I get
11:59:47mcuelenaeredoes anyone have a clue?
12:00:52LearAnyone know how accurate the PC is for a data abort exception on ARM?
12:03:32mcuelenaerewhich program sets up sysfont.h?
12:06:31mcuelenaereI meant the actual program, I believe make just calls it?
12:08:07JdGordonsorry, yeah
12:10:02JdGordon$(BUILDDIR)/sysfont.h: ../fonts/rockbox_default.bdf
12:10:02JdGordon$(call PRINTS,Create sysfont.h)$(TOOLSDIR)/convbdf -h -o $@ $<
12:10:15JdGordonin frimware/Makefile
12:10:25mcuelenaerethanks, I was looking build/Makefile
12:12:53 Quit K4rP4D ("Leaving")
12:12:54 Quit mcuelenaere (Read error: 131 (Connection reset by peer))
12:17:14 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
12:17:21mcuelenaereis there a special option to set for making sysfont.h?
12:18:12B4gderuse make V=1 to see the full command lines
12:21:16*GodEater continues fighting with the O2 upgrade site
12:21:45mcuelenaereit just omits the creating of sysfont.h for some reason .
12:23:32 Join culture [0] (
12:27:46 Quit mcuelenaere (Read error: 131 (Connection reset by peer))
12:32:50 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
12:38:57markunmcuelenaere: the sysfont.h error is usually because of something else
12:38:59 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
12:43:56 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
12:43:58***Saving seen data "./dancer.seen"
12:51:38 Quit safetydan ("Leaving")
12:53:05 Quit mcuelenaere (Read error: 54 (Connection reset by peer))
13:00:14mrkikomcuelenaere, markun said for you "the sysfont.h error is usually because of something else"
13:00:22mrkikoI told you so cause I saw you disconnected
13:00:51 Quit mrkiko ("SIGSEGV")
13:00:53 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
13:03:27markunmcuelenaere: try this:
13:03:54markunmaybe it will tell you where the real problem is
13:04:55B4gderyes, that error is usually a sign of SOURCES having a file mentioned that doesn't exist
13:05:31B4gderannoying that we have no better way to detect that yet
13:08:51JdGordonatm, the ata callbacks are called every 3 sec on flash targets... does anyone have objections with changing that to somwhere between 60 and 120s (or higher even?)
13:08:51 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
13:09:13JdGordonNico_P: that would make it work with buffering more nicely yeah?
13:10:05Nico_PJdGordon: buffering doesn't use the callback ATM, but it would probably be better for when we start using it again, yeah
13:13:44 Join bertrik [0] (n=5a80bda4@gateway/web/cgi-irc/
13:15:04 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
13:17:33 Join Shaid [0] (
13:17:41mcuelenaeremarkun, B4gder: thanks
13:18:30JdGordonhas anyone had any crashes when pressing a button with the backlight off recently?
13:19:43JdGordondid you ever have that problem?
13:19:50mcuelenaereJdGordon: are you talking about the D2?
13:19:59JdGordonno, in general
13:20:03 Join faemir [0] (
13:20:06mcuelenaereah no, never mind then
13:20:27JdGordon7655... but I've had in on my e200, but AGES ago...
13:21:57mcuelenaereis it possible to set #error or #warning in SOURCES?
13:23:46B4gderyes, but I don't think you'll see that output thoughe
13:24:14 Join Thundercloud [0] (
13:26:24JdGordonB4gder: out of curiosity... about how many have been nominated (and acepted) for the rsb?
13:26:47B4gderwe're at ~10 accepted
13:27:49LearNico_P: Should fill_buffer lock the list mutex somehow?
13:28:21 Quit mcuelenaere (""setting up router"")
13:30:01JdGordonanyone want to check the polish lang update? or should i just commit?
13:30:25Nico_PLear: IIRC, no, but let me check
13:36:05markunJdGordon: I also wonder about that all the time. Should we have committers who speak the languages?
13:36:13Nico_PLear: the mutex is intended to protect the state of the linked list... fill_buffer doesn't do much on it
13:36:50markunI was happy that TiMiD committed the Japanese file, since there was some confusion.
13:37:46markun(2 translations and one guy who keeps interfering with langauges he doesn't speak)
13:38:06JdGordonmarkun: yeah, would be nice if commiters would look out for langs they speak
13:38:11JdGordonand yes, that guy was bloody annoying
13:38:14LearNico_P: It does iterate over it, while calling code that could yield. Not too different from when loading album art?
13:38:49 Quit Shaid (Read error: 110 (Connection timed out))
13:39:33LearThing is, under very specific circumstances, I get data aborts in buffering.c...
13:41:57Nico_Pwhat kind of circumstances?
13:43:16 Join LambdaCalculus37 [0] (i=44a0430f@gateway/web/ajax/
13:45:12 Join cg_ [0] (n=cromos@
13:45:12 Quit cg (Read error: 104 (Connection reset by peer))
13:47:59LearToo long to tell, almost. :) On e200, when resuming (or playing?) certain tracks using album art and dircache (never without dircache).
13:48:30 Quit nplus (Read error: 104 (Connection reset by peer))
13:49:29Nico_Phave you every tried it on the sim? also, have you traced the data abort to a particular function?
13:51:30LearNever seen it in the sim, but I haven't tried all that much. Exact function can vary between builds at least, but I've now had two different aborts (in different functions), both caused by a bad next pointer.
13:53:36LearNo, same function, but different place within it.
13:54:06 Quit PaulJam (Read error: 113 (No route to host))
13:55:38 Join Arathis [0] (
13:55:51 Join PaulJam [0] (
13:56:28 Quit ivor ("Leaving")
13:57:05Nico_Pyou're probably right about the need for the mutex
14:00:44 Join potato [0] (
14:01:11LearTried adding a call to find_handle after buffer_handle (in fill_buffer), but that only changed when the data abort happened.
14:08:44 Join n1s [0] (n=nils@rockbox/developer/n1s)
14:14:48 Quit bertrik ("CGI:IRC (EOF)")
14:16:37 Join miepchen^schlaf [0] (
14:17:53 Part potato
14:25:38 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
14:31:03 Join Vorador [0] (
14:37:59 Join nplus [0] (
14:40:28 Join Schmogel [0] (
14:43:59***Saving seen data "./dancer.seen"
14:45:36 Join MarcGuay [0] (
14:47:31MarcGuayrasher: Thanks for updating the fixme's page.
15:00:03 Part B4gder
15:00:07 Quit perrikwp (" ajax IRC Client")
15:01:19 Join webguest46 [0] (n=42e39f64@gateway/web/cgi-irc/
15:04:56 Join ctaylorr [0] (
15:06:59 Join Arathis2 [0] (
15:07:04webguest46I've been trying to find out if there have been any projects started at developing rockbox for a Zune player. I haven't found any mention of the idea anywhere and am very curious of the possiblity. Any info would be appreciated.
15:07:50desowinthere was quite a lot people saying about that, but no code I know of showed up
15:07:52GodEaterwebguest46: you can't have looked too hard, I'm sure there's a thread in the NewPorts section on the forums, let me go find it
15:08:58webguest46TYTY.... i guess i missed it....
15:09:28 Quit CriminalScum (Client Quit)
15:11:43 Quit blithe ("Be right back.")
15:11:55 Join blithe [0] (
15:12:26 Quit blithe (Client Quit)
15:12:43 Join blithe [0] (
15:16:30 Quit amiconn (Nick collision from services.)
15:16:37 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
15:21:17 Quit k4y (Remote closed the connection)
15:22:24 Join k4y [0] (
15:24:01 Quit webguest46 ("CGI:IRC (EOF)")
15:24:38 Quit Arathis (Read error: 110 (Connection timed out))
15:24:56 Join stoffel [0] (
15:26:36 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
15:27:24MarcGuayWhat's difference between the X5V and the X5? Is this something your average user can tell without further guidance?
15:28:45LambdaCalculus37IIRC the X5V doesn't have the FM radio.
15:29:27LambdaCalculus37I don't know if there's an easy way to tell on the outside, though.
15:29:38MarcGuayI'm wondering if "Use the x5v_fw.bin file if your player is a X5V. If it is a X5, use the x5_fw.bin file." in the installation instructions is sufficient...
15:30:21LambdaCalculus37Personally, I'm certain that the owners of the device should know exactly which model they have. But I don't own an iAUDIO X5, so I can't really say for sure.
15:30:46LambdaCalculus37I'm *assuming* that "X5V" must be printed on the back of the device, as Cowon did that with the X5L.
15:31:20LambdaCalculus37You could change it slightly to read "If it is an X5 or X5L, use the x5_fw.bin file", though.
15:37:51MarcGuayDoes the iAudio firmware actually get erased when rockbox is installed or can the bootloader just not point towards it?
15:38:28LambdaCalculus37I think the bootloader simply doesn't point to it.
15:38:41LambdaCalculus37But I think we should consult an X5 dev on this.
15:38:43MarcGuayThe manual says " It replaces the original firmware", which seems wrong...
15:39:00 Join K4rP4D [0] (n=KrPD@unaffiliated/krpd)
15:39:10LambdaCalculus37Well, I do know for a fact that the X5 officially does not dual boot.
15:39:27*LambdaCalculus37 casts the "Summon an X5 dev" spell
15:39:39MarcGuayI think it means to say "The bootloader replaces the OF in the boot sequence with Rockbox."
15:40:01*MarcGuay isn't convinced that's any clearer.
15:40:15LambdaCalculus37MarcGuay: But it gets the idea across better.
15:40:28MarcGuayIf it's true. :) Guess I'll wait...
15:40:33LambdaCalculus37Actually, no... scratch that.
15:41:06LambdaCalculus37Like I said, the X5 doesn't dual boot. Ditto the Gigabeat F. So in those cases, Rockbox is pretty much replacing the firmware.
15:44:00MarcGuayI think I understand now - Because the device has a hardware bootloader, the rockbox bootloader is written over the OF.
15:47:01MarcGuayIs there a reason there's no advice on how to backup the OF in the X5/M5 instructions before overwriting it, as is done for the Gigabeat?
15:48:26peturI thought there was a new, working dual-boot loader for x5?
15:48:46LambdaCalculus37MarcGuay: The biggest reason why it's done for the Gigabeat is because the firmware actually resides on the hard drive, so if you ever want to go back, you should have an unaltered backup of the OF.
15:49:13LambdaCalculus37petur: There was a patch on Flyspray, but I never saw word of an official one.
15:49:51MarcGuayLambdaCalculus37: The instructions for the iAudio suggest that you go to their website and download the firmware again to uninstall. Kind of weak, no?
15:50:08MarcGuaypetur: No clue on that, just going by what I've read.
15:50:15peturI was told on devcon that the new one was looking great and could be made official after enough testing
15:50:53LambdaCalculus37MarcGuay: Agreed.
15:51:19 Quit ctaylorr (Read error: 110 (Connection timed out))
15:51:35MarcGuayI feel like I'm stumbling around in the dark here.
15:52:01 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
15:52:51MarcGuayIf no one speaks up I'm going to reword it to "The Rockbox bootloader overwrites the original firmware, making it impossible to dual-boot." and move on...
15:53:25n1sMarcGuay: IIUC you simply can not back up the iaudio firmware as there is no way to get it from the flash
15:54:12MarcGuayn1s: Nice, thank you... And we can't provide a copy of it because of copyright?
15:54:21n1si guess so...
15:55:38n1sOk, so one week with 0 useful comments in the tracker should mean no one objects to me committing keybox, right (I will of course provide a manual update)
15:56:08 Join jgarvey [0] (
15:57:17*JdGordon objects!
15:57:35JdGordonyou cant go round commiting stuff with manual updates!
15:57:40JdGordonits just not done!
15:57:45JdGordonthink of the children!!!
15:58:26MarcGuayJdGordon: Your horns are showing.
15:58:45*LambdaCalculus37 files off JdGordon's horns
15:58:46JdGordon\M/ o_0 \M/
15:59:56*n1s interprets JdGordon's reply as: "please go ahead" :)
16:02:23MarcGuayWhat's the story with recording on the ipods again?
16:02:48n1splayback will crash/do bad things if you try to start it after recording
16:03:02JdGordonnot on all of them though
16:03:20LambdaCalculus37It does break things on the iPod video and 4G color.
16:03:48MarcGuayBut they *can* record? I don't recall seeing the menu on the nano (don't have it with me).
16:04:11JdGordonthey need an adaptor for the dock dont they?
16:04:27 Join MethoS- [0] (
16:04:32MarcGuayJdGordon: That's it, thanks.
16:04:46JdGordondont take my word for it though...
16:04:47 Quit K4rP4D ("Leaving")
16:05:01n1sbut the nano can't record because the dac doesn't support it
16:05:46n1summ, ignore me
16:06:01JdGordon... was about to say.. not according to its config
16:06:11 Quit faemir (Read error: 104 (Connection reset by peer))
16:06:16 Join arohtar [0] (
16:06:44n1si mixed it up with some other target, possibly the mini
16:07:15MarcGuayOkay, I give up for today... If anyone wants to check it out I was trying to fix up the recording keymaps in the ipods w/ recording manuals.
16:09:03preglowgreat, another *box plugin
16:09:39JdGordonat least it is a "box"
16:17:23GodEaterhas jhMikes been in recently ?
16:19:24n1sBagder: build seems to have stuck :/
16:26:59 Quit stoffel ("leaving")
16:31:38 Quit DataGhost (Nick collision from services.)
16:31:46 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
16:33:57 Join WakeUpWolfgang [0] (n=433797c0@gateway/web/cgi-irc/
16:34:38WakeUpWolfgangi have a question
16:35:09WakeUpWolfganghow can i get videos on my ipod nano
16:36:17GodEatercopy them there ?
16:36:24PaulJamWakeUpWolfgang: convert them to the correct format (see PluginMPEGPlayer wiki page) and copy them somewhere on your player.
16:36:28GodEaterassuming you have them in the right format
16:37:33WakeUpWolfgangok thanks this is my frist time useing rockbox
16:39:17PaulJamsorry, the page is named "PluginMpegplayer"
16:42:07LearNico_P: How is shrink_buffer supposed to work, really? :)
16:43:22LearLooks kind of dangerous to me, especially with it first "buffering up" memory handles, and then start moving them about...
16:44:03***Saving seen data "./dancer.seen"
16:44:20Nico_Pyou mean in the Q_START_FILL event?
16:46:52Nico_Pthe shrinkage happens before the buffering is done
16:47:35Nico_Pthough I'm tired so I might be misreading the code
16:51:21 Quit WakeUpWolfgang ("CGI:IRC 0.5.9 (2006/06/06)")
16:55:16LearSo that's the code responsible for freeing up old data, to make room for new stuff?
16:57:49LearIf I disable that code (maybe a bad idea as such), I can't reproduce the data aborts...
17:00:35 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
17:01:51 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
17:09:44Nico_PLear: interesting... and yes that code frees up old data
17:12:56 Join MethoS_ [0] (
17:13:22*n1s pokes the buildserver with a long stick
17:14:35 Join MethoS-- [0] (
17:14:50MarcGuayThe 3rd and 4th gen ipods can only record through the dock, correct? Apple never produced a unit with recording built-in, did they?
17:16:29LambdaCalculus37Nope, they never did.
17:16:45LambdaCalculus37All the recording methods for the iPod (that I've seen) have all been third-party.
17:17:06 Quit n1s ()
17:18:14 Join delorean90 [0] (
17:18:38 Join Horschti [0] (
17:18:43PaulJami thought there was one generation that allowed recording through the headphone socket.
17:18:53GodEater3rd gen allowed it
17:19:03MarcGuayLambdaCalculus37: Thanks.. though it looks like they sold a microphone that went in through the headphone jack...
17:19:04LearNico_P: It seems like the problem is that if the bufopen for album art is called too quickly after the bufopen for the metadata (which triggers the start fill), then I can get the abort.
17:19:24 Quit Horscht (Nick collision from services.)
17:20:19MarcGuayLambdaCalculus37: Does the 4th gen have a recording option on the main rockbox menu?
17:20:36LambdaCalculus37MarcGuay: Let me check again.
17:21:24MarcGuayLambdaCalculus37: Ah, I finally see the recscreen in keymap-ipod.c. Thanks folks.
17:22:06MarcGuayLambdaCalculus37: Can you pull up the recording menu from within the recording screen?
17:22:48Nico_PLear: have you tried adding mutex locking to fill_buffer?
17:23:52MarcGuayLambdaCalculus37: Looks like the menu button should do it. Long Play/Pause to exit.
17:25:13LearNico_P: No, not yet. Should I do it around the entire loop?
17:26:39LambdaCalculus37MarcGuay: Pressing menu does nothing in the recording screen.
17:26:57LambdaCalculus37Actually, it just jumps back to the root menu.
17:27:13LambdaCalculus37There's no way of getting into the Recording Settings menu from within the recording screen.
17:27:41Nico_PLear: that won't work well because it will mean that a handle can't be added until the buffering is finished... I think it's a bit trickier than that
17:27:50LearNico_P: Need to re-get m after calling shrink_handle in fill_buffer at least...
17:27:55LearSuspected as much.
17:27:57Nico_Pi'll try to have a look when I get home
17:29:09 Quit MethoS- (Read error: 110 (Connection timed out))
17:29:13 Join MethoS- [0] (
17:30:19Nico_Pmaybe protecting the shrink_buffer call in fill_buffer would be enough...
17:32:19 Quit MethoS_ (Read error: 110 (Connection timed out))
17:32:55Learmove_handle grabs the mutex, so shrink_handle should be fine. However, m might have changed.
17:36:01MarcGuayLambdaCalculus37: Can you get to the quickscreen from the recscreen with a long-menu?
17:36:48LearHm, seem to have something that works now. Maybe overdoing it a bit though. :) Should I make a patch for it?
17:36:53 Join LambdaCalculus3 [0] (i=44a0430f@gateway/web/ajax/
17:37:06Nico_PLear: sure :)
17:37:26 Quit LambdaCalculus37 (" ajax IRC Client")
17:37:38 Nick LambdaCalculus3 is now known as LambdaCalculus37 (i=44a0430f@gateway/web/ajax/
17:37:39Nico_PLear: shrink_handle should probably take a struct memory_handle ** as arg, and update it
17:38:40MarcGuayLambdaCalculus37: Forget that, sorry, I just found the problem.
17:38:57LearNico_P: Yes, that thought crossed by mind too... Only really needed in one place though.
17:41:44 Join ch4os [0] (n=ch4os@unaffiliated/ch4os/x-059673)
17:42:14 Join mf0102 [0] (n=michi@
17:43:37 Quit MethoS-- (Connection timed out)
17:43:52Nico_PLear: I'm heading home. I'll check the logs for your patch ASAP
17:44:32 Quit Nico_P (" ajax IRC Client")
17:45:45 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
17:47:29 Join MethoS_ [0] (
17:50:27 Quit BigBambi (" ajax IRC Client")
17:52:29MarcGuayLambdaCalculus37: Would a short or long select make the most sense to bring up the recording menu from the rec. screen? (or something else completely?)
17:54:17LearNico_P: Added as FS #9164.
17:55:02LambdaCalculus37MarcGuay: I say long select, as it's consistent with bringing up the quick menus.
17:56:26 Join mcuelenaere_ [0] (n=mcuelena@rockbox/developer/mcuelenaere)
17:57:03 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
17:58:18 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
17:59:35MarcGuayLambdaCalculus37: Care to test a patch?
18:00:00 Join midgey [0] (
18:00:32 Join elvalecita [0] (
18:01:59LambdaCalculus37MarcGuay: I don't have my laptop today. Can you compile a build for me?
18:02:09LambdaCalculus37Also, apply FS #9122 to it.
18:03:19 Quit MethoS- (Read error: 113 (No route to host))
18:03:28 Join herrwaldo [0] (
18:05:40 Quit petur ("work->home")
18:08:57 Nick Lear is now known as Lear_ (i=chatzill@rockbox/developer/lear)
18:09:36 Quit mcuelenaere (Read error: 113 (No route to host))
18:14:13 Join Mathiasdm [0] (
18:14:37 Quit mcuelenaere_ (Read error: 60 (Operation timed out))
18:16:25 Join dany_21a [0] (n=dan@
18:16:36 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
18:16:46 Quit HellDragon (Read error: 104 (Connection reset by peer))
18:18:24 Quit Genre9mp3 ()
18:19:32 Join bertrik [0] (n=5a80bda4@gateway/web/cgi-irc/
18:23:57elvalecitahello! im new with rockbox, fresh installed in my ipod 5.5 gen, 30 gb, running win xp sp2. I have a lot of playlists and I converted them using "iPod-iTunes to RockBox Playlist Converter v1.01", the developer explains that I should use a patch so the songnames are displayed correctly (FS #7652 - EXTM3U support). I have been trying to learn how to actually install this patch.
18:24:03elvalecitaAfter reading the guide in the wiki ("WorkingWithPatches"), I guess its too hard for me being no programmer. Is there an less complicated way to fix the display of the song names? Should I give up and convert the lists manually (would take pretty long)? Thanks for your attention! Javier
18:25:45 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:26:59 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
18:28:27pixelmaMarcGuay: I remember having a look at the recording keymaps of the Ipods once (for the manual too) and it wasn't fully working
18:29:12Lloreanelvalecita: The problem is that the songs were synced with iTunes. It renames the song files. The EXTM3U reads extra lines with the song names in the playlist file, that your program put there.
18:29:19LloreanThe only other fix is to simply have properly named songs.
18:32:55pixelmalooking at the diff you pasted - long select once was an idea for my c200 keymap changes (in the recording screen) but I scrapped that since as long as a short select starts a new recording it's far too easy to accidentally start recording if you would just like to enter the menu...
18:33:10pixelma^ MarcGuay
18:33:13elvalecitaLlorean: well i've actually used Winamp the whole time, i guess it does the same that itunes. So what you are saying is that I should erase my whole library from the ipod and then copy it from the original files?
18:33:24 Join MethoS- [0] (
18:33:50 Join HellDragon [0] (
18:33:51 Join stoffel [0] (
18:34:05Lloreanelvalecita: That would be the easiest way, yes.
18:34:16LloreanDon't treat it like an iPod. Treat it like an external disk.
18:34:40elvalecitai had kind of hoped that my existings lists would get lost
18:34:55 Part pondlife
18:34:58elvalecitabut i guess it is quicker to redo them than to take a programmer course
18:35:13LloreanLearning to compile and patch doesn't actually take any programming knowledge.
18:35:22LloreanIt's just a few instructions to follow.
18:35:38 Quit EspeonEefi ("さよなら")
18:35:53LloreanBut if the patch is out of date, that can be a problem. Plus, we don't support modified versions. And so on.
18:36:26 Join MethoS-- [0] (
18:36:45elvalecitaso if I could patch it, i couldnt update the program?
18:37:44LloreanYou could update it. You'd have to compile the updates yourself (not very hard at all0
18:38:03LloreanBut if you had any problems, the first thing we'd do is ask you to install an unpatched version and make sure they happen without the patch
18:38:58mcuelenaeremm for some reason target/mips/* becomes target/1/* in the Makefile, any ideas?
18:38:59elvalecitaso it would work if i follow the instructions in the wiki?
18:40:18Lloreanelvalecita: If the patch is up to date. If it doesn't work, then yes, you'll have to do a few more complicated things
18:40:42 Join miepchen^schlaf_ [0] (
18:41:11elvalecitaso i would have to do the Applying A Patch
18:41:11elvalecitaApplying a 'patch' (output from diff -u) is done with the 'patch' tool:
18:41:31elvalecitaonly the "applyng"?
18:42:19LloreanNo, you have to compile it afterward.
18:43:17elvalecitaok, compiling a C program its not that hard, i guess.
18:43:27GodEater_anyone else read the "Quandry" email on the mentor's list, and then follow the link to the "bad patterns" link in it ?
18:43:34GodEater_made for quite interesting reading I thought
18:44:07***Saving seen data "./dancer.seen"
18:44:12 Join Galaxor [0] (
18:44:15Lloreanelvalecita: No, compiling itself is just typing out a few commands. As long as you follow the directions you don't really need to learn anything. :)
18:44:18*GodEater_ thinks he used the word "link" there too often
18:44:18elvalecitaWell i have to go, but many many thanks for the help. ill try the applying and compiling thingie and if doesnt work ill convert the lists. Godd evening (or day) to you all.
18:44:22PaulJamelvalecita: if you are on windows i would recommend the SimpleGuideToCompiling linked from here:
18:45:58 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
18:46:41LloreanGodEater_: How many times have we seen someone say "I just want to polish it up some, then I'll release the patch" etc?
18:46:43GalaxorHi. I got the latest build as of Saturday, for sansa c200. It's highly excellent, with one exception: When I plug into the usb, it reboots me into the old firmware. Is that a bug, or have I somehow misconfigured it?
18:46:43elvalecitaok ill check it out PaulJam
18:46:49GodEater_Llorean: exactly
18:46:55GodEater_I was thinking "blackhawk" in particular
18:47:12GodEater_Galaxor: that's intended behaviour currently
18:47:23LambdaCalculus37Galaxor: Rockbox doesn't have a full USB stack yet.
18:47:23 Quit elvalecita ()
18:47:25 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
18:47:37GalaxorGodeater: Oh, okay.
18:47:49GodEater_if it didn't do that - you'd have no USB access to it
18:48:07LloreanGodEater_: Indeed, but just one of many. There's a popular Album Art tool people use with Rockbox, coded up by someone at the ABi forums, and it took a good deal of work to talk him into releasing his code. He wanted to, but he also wanted to "clean it up" first.
18:48:18GodEater_ah yes, I've seen it
18:48:28GalaxorThis is kernel stuff?
18:48:38GodEater_Galaxor: usb you mean ?
18:48:42GalaxorThe usb support, yeah.
18:48:48GodEater_yes, it's not in there yet
18:48:53GodEater_not in normal builds anyway
18:49:11GalaxorOkay. I see, thanks.
18:49:35GodEater_the plan is to put it in eventually, but it has quirks at the moment which make it undesirable for "normal" users
18:49:44GalaxorWhat kind of quirks?
18:49:51GodEater_file system corruption
18:50:03GalaxorThat does seem like an undesirable quirk.
18:50:14GodEater_aye ;)
18:50:38GodEater_gevaerts here is the expert
18:50:42GodEater_if you want more detail
18:50:53 Quit MethoS_ (Read error: 110 (Connection timed out))
18:50:54 Join OlivierBorowski [0] (
18:50:56*GodEater_ drops gevaerts in it
18:51:03 Quit miepchen^schlaf (Connection timed out)
18:51:12 Quit MethoS- (Read error: 110 (Connection timed out))
18:51:13LambdaCalculus37MarcGuay: Still here?
18:51:13*mcuelenaere really hates these massive #ifdef's files
18:51:34*gevaerts thinks that it is unfair to say that the file system corruption issue is a USB problem
18:51:58GalaxorThis stuff is under firmware/usbstack?
18:52:00LambdaCalculus37gevaerts: It's more of a "feature". :P
18:52:20LloreanLambdaCalculus37: It's more "an existing problem that USB brought to light" if evidence holds.
18:53:11gevaertsAnyway, there are a few issues left : "something" timing related that shows up more often when using hubs, and the (mentioned) corruption issue, which is probably a bug in the sansa sd driver
18:53:11GalaxorOoh, I see the usb_charging_only. Is that a thing where you can plug in the machine and it will charge but not act like a disk, and then I can reboot the machine manually in order to connect the player to the computer?
18:53:47GodEater_Galaxor: yes, the manual should say which key to hold while plugging in USB does that for you
18:54:11 Quit MethoS-- (Connection timed out)
18:54:18gevaerts(there is a third issue, but that only applies to gigabeat S and mac osx)
18:54:32GalaxorOh, okay. I can use that for now. I might poke my head in the code, but I've never done kernel-level stuff before, so I'll probably not be useful there.
18:54:57GodEater_gevaerts: what's the third issue then ?
18:55:04*GodEater_ doesn't recall any issues on his GBS yet
18:55:10mcuelenaereanyone around who's familiar with the Rockbox Makefile and/or SOURCES file parsing?
18:55:33 Quit nplus (Read error: 104 (Connection reset by peer))
18:56:21gevaertsGodEater_: I could be wrong, but I seem to remember that some versions of OSX don't like it if a device that claims to be non-removable doesn't implement the "caching" mode sense page
18:56:57gevaertsGodEater_: this matters for the GBS because it has two FAT partitions, and Windows doesn't handle more than one partition on removable devices
18:57:17GodEater_gevaerts: it doesn't? Since when ?
18:57:19*gevaerts thinks that both behaviours are bugs, but there's not much we can do
18:57:38*GodEater_ has plugged in multi-partition USB keys to his PC before
18:57:57GodEater_although I guess only one of them was a FS that windows understood
18:58:01GodEater_so perhaps that made a difference
18:58:26MarcGuayLambdaCalculus37, pixelma: I have company over. Will check it out more later, and that's a good point about the "new recording" button.
18:58:35gevaertsIt does. s/more than one partition/more than one Windows-accessible partition/
18:58:44LambdaCalculus37GodEater_: I thought both partitions were FAT32, and it was just a flag set to the system partition (according to the wiki, anyway).
18:58:52LambdaCalculus37MarcGuay: Okay, no problem.
18:59:04GodEater_LambdaCalculus37: they're TFAT technically
18:59:17GodEater_and I've never tried plugging it into a windows PC, so I've not seen any weird behaviour
18:59:20Lloreangevaerts: Doesn't the e200 register as a removable device with two partitions (internal flash and SD?)
18:59:22*gevaerts wonders what the T stands for. "Terrible" ?
18:59:30LloreanOr are they treated as two separate devices?
18:59:33GodEater_that would be two volumes
18:59:36GodEater_not partitions
18:59:38LloreanAh, right.
18:59:40gevaertsLlorean: they are two LUNs
18:59:42GodEater_T = Transaction
18:59:53LloreanCan that be faked?
19:00:06GodEater_I would imagine so
19:00:12LloreanWe don't exactly want people repartitioning their S anyway
19:00:13GodEater_means more work for gevaerts though ;)
19:00:45gevaertsLlorean: maybe not, but we may want them to be able to make a full disk-level backup
19:00:52GodEater_the S wouldn't want them repartioning it either
19:00:55LambdaCalculus37My S shows up as two drives, plus a USB Camera Device (which is odd) on Kubuntu.
19:01:00LloreanGodEater_: I think that's why we don't. :)
19:02:14*gevaerts is willing to make a translation layer to allow this sort of trick, but he won't be tricked into deciding that it is needed
19:03:38gevaertsLambdaCalculus37: I deny all responibility for the camera
19:05:28GodEater_gevaerts: leave it to RSB ?
19:05:52*Llorean begins to fear how much is going to get dumped on the RSB.
19:06:12gevaertsGodEater_: maybe. There are good arguments for both sides
19:06:39GodEater_gevaerts: it won't be popular, but I guess we could make it an option ?
19:06:49GodEater_i.e. some sort of toggle somewhere
19:07:02gevaertsWouldn't that be a support nightmare?
19:07:11GodEater_it could well become one
19:07:23GodEater_but then, *not* having the translation layer might well become a support nightmare
19:07:31GodEater_a lot of our "users" do run windows after all
19:07:39LloreanI think that turning off the translation layer could be hidden in debug
19:07:42GodEater_if the GBS become a supported target
19:07:44Lloreanno "userspace" toggle.
19:07:54gevaertsAnd not persistent
19:08:01GodEater_that's a good idea
19:08:01Llorean"debug" rarely is persistant
19:08:04*gevaerts could live with that sort of option
19:08:06LloreanIt doesn't save in .cfg files
19:08:17GodEater_e.g. usb screenshot mode
19:08:19 Join OlivierBorowski_ [0] (
19:08:22gevaertsBut I'm still in favour of no translation layer
19:08:44LloreanWould the translation layer mean more overall compatibility?
19:08:46 Join shadoxx [0] (
19:08:52LloreanLess "unexpected" behaviour for the common user?
19:09:12shadoxxAnyone familiar or good at identifying ICs?
19:09:17GodEater_gevaerts: well do you have another idea for getting this to work on windows then ?
19:09:31 Join petur [50] (n=petur@rockbox/developer/petur)
19:09:42gevaertsGodEater_: it works now. It's just a bit unsafe.
19:09:51GodEater_unsafe how ?
19:10:10gevaertsYou can access the firmware partition and the partition table
19:11:17gevaertsAnyway, if the consensus (or RSB decision if there isn't one) is that there should be a translation layer, I won't oppose it
19:11:23GodEater_how is that unsafe ? Isn't that normal behaviour ?
19:12:01gevaertsIf it isn't considered unsafe, then why are we discussing measures to make that harder?
19:12:10*gevaerts is a bit confused now
19:12:38GodEater_you said "windows doesn't handle more than one parition on a removable drive"
19:12:40 Quit Thundercloud (Remote closed the connection)
19:12:49GodEater_I assumed that meant it only allowed access to one of them
19:12:49LloreanGodEater_: Right now it's not "removable"
19:13:05GodEater_oh - how is it marked currently then ?
19:13:11 Join Thundercloud [0] (
19:13:11gevaertsRemovable is a SCSI term with a precise meaning :)
19:13:24*GodEater_ is clearly not the right person to be talking about this given he doesn't use windows with his S
19:13:27LloreanIt's removable, but not "Removable"
19:13:30gevaertsNothing to do with the USB removability
19:13:52LloreanI think access to the partition table is probably "bad" but access to the firmware partition is good (for easy updates)
19:13:55GodEater_ok - I think I'll butt out now then while I'm only vaguely confused :)
19:14:20gevaertsSCSI removable means that the SCSI handler stays there, but the media can get removed (floppy, SD card,...)
19:14:43gevaertsOnly a lot of USB device manufacturers didn't read that properly
19:15:06GodEater_that doesn't sound like desirable behaviour for a DAP without an SD card slot
19:15:32GodEater_I can see it would make sense for the sansas though
19:16:00gevaertsIt's a bit complicated, because for lots of reasons that bit also gets used by OSes to make decisions, like "do I put a trashcan on it"
19:16:07 Quit Vorador ("Ex-Chat")
19:16:24gevaertsSo for this bit too, there is no "right" answer
19:16:29 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
19:17:46GodEater_how irritating
19:17:49gevaertsA useful thing you get from setting the removable bit is that you can get eject commands, which you can use to leave disk mode and go to charging mode (technically you are still in disk mode, but the disk is ejected)
19:18:05GodEater_which is presumably how the ipod does it
19:18:14*Llorean would very much like that.
19:18:26*Llorean also does like the "doesn't keep a trashcan" aspect of removable devices.
19:18:34*gevaerts points out that it already works like that, except for the GBS
19:20:50gevaertsYou can even go back to disk mode, via the debug menu
19:21:25*gevaerts wonders when Bagder will be able to kick the build master
19:22:08 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
19:22:24LloreanOh, right, I remember that now from when I tried on my Nano
19:25:34 Quit OlivierBorowski (Read error: 110 (Connection timed out))
19:27:11 Quit culture (Connection timed out)
19:32:29 Quit vort3x (Read error: 104 (Connection reset by peer))
19:34:21preglowi hate whatever way flyspray uses to post replies that has me posting another one each time i reload a page
19:35:31GodEater_GET and not POST is usually the culprit there
19:36:15preglowyeah, probably
19:39:15 Join stripwax [0] (
19:39:33 Quit OlivierBorowski_ (Remote closed the connection)
19:39:50 Join OlivierBorowski_ [0] (
19:40:01dany_21ahello all... i am daniel_at from the "Sansa V2"-thread from the rb-forum... i have some ARM (and ARM-assembly) related questions, anyone here who can give me some hints?
19:40:19GodEater_dany_21a: I'm guessing amiconn can probably help
19:40:29LambdaCalculus37dany_21a: Ask away.
19:40:29GodEater_assuming he has time
19:41:42dany_21aLambdaCalculus37: okay... thx: i have none to nearly none ARM-assembly knowledge (once or twice debugged a Pocked-PC programm... thats all) so I have some basic questions:
19:41:59dany_21aWhat does "LDR R0 [R0,#8]" to
19:42:09dany_21asry... What does "LDR R0, [R0,#8]" to
19:42:32GodEater_that looks like the same question ;)
19:42:36dany_21a(What does the [...]-bracket thingy do)
19:42:37*amiconn multicast-pings LinusN, Bagder and Zagor
19:42:48dany_21aGodEater: (forgot semicolon)
19:43:13GodEater_I see no semicolon...
19:43:30dany_21aahm... whats the english word for ","
19:43:40dany_21aokay..forgot that :)
19:43:47GodEater_semi-colon -> ;
19:44:06dany_21aokay.. thx
19:44:16amiconndany_21a: It loads a 32-bit value from address (r0 + 8) into r0
19:44:26*amiconn recommends the arm reference manual
19:44:51dany_21aamiconn: so [..] is basically the offset-operator?
19:45:34amiconnIt's not an operator, it's an addressing mode. All ldr and str instructions use []
19:46:06amiconnGoogle for ddi0100e_arm_arm.pdf
19:46:14 Quit stoffel ("leaving")
19:48:26dany_21aamiconn okay, will have a look
19:50:24 Join vort3x [0] (n=vortex@unaffiliated/dfa001)
19:53:49 Join tvelocity [0] (
19:53:57dany_21aamiconn: does this make sense:
19:55:45delorean90i noticed in the whyrockbox section of the wiki that While Playing screen was in bold, but only the first 2 words were capitalized, i just got permission for the wiki yesterday, should i capitalize the 's'?
19:55:56dany_21auhmm.. just saw.. the ands should read "ands r1, r1, #0x02" - or?
19:56:00 Quit ze (Remote closed the connection)
19:56:04 Join ze [0] (
19:56:13dany_21aamiconn: ^
19:56:20amiconn(1) You cannot use ldr(b|h) with = this way. It will load the address itself from the constant pool (but that cannot work because ldrb is byte sized)
19:56:29amiconn(2) There is no 'noloop' label
19:56:51amiconnerr, scratch (2)
19:56:56dany_21a(last line)
19:57:22dany_21ato (1)... okay - how to i load the value (its actually a register) from a constant address.
19:57:27amiconnyeah, just a bit far away to spot it easily...
19:58:04amiconnYou need to load that address into a register first. Either using ldr, from the constant pool, or using mov/orr sequences with immediates
19:58:42amiconnAnd there's 'tst' for testing bits, which does the same as 'and' but discards the result itself
20:01:03 Quit bertrik ("CGI:IRC (Ping timeout)")
20:02:17amiconnThe other version would be to use mov r1, #0xC8000000; orr r1, r1, #0x0D0000 instead of ldr r1, =GPIO3_DATA
20:03:19 Quit itcheg ("Life without danger is a waste of oxygen")
20:03:40dany_21aamiconn: okay - i have already seen that construct from the pastpin in the disassmbly some times
20:03:44amiconnThis uses the same amount of binary size (the ldr also needs the address in the constant pool, automatically put there by the assembler). Whether it's faster or slower depends on the arm version and generation
20:03:52dany_21athx - currently compiling/uploading
20:04:14amiconnOn arm7tdmi the mov/orr combo is faster. On later ones it probably doesn't matter
20:05:12dany_21aamiconn: yeah, just wondered how <shift_operand> can be imm32
20:05:41dany_21a"[...](the ldr also needs the address in the constant pool,[,...]"
20:06:05amiconnYes, that means it's outside the instruction flow
20:07:11 Quit Acky (Read error: 104 (Connection reset by peer))
20:07:54dany_21aamiconn: just to be sure... the loop is executed if the bit1 is set...or?
20:08:48amiconnThe assembler collects everything you assign using =, and dumps this constant pool somewhere within range for offest addressing, either automatically at the end of the module, or wherever you put a .ltorg directive
20:09:16 Nick Lear_ is now known as Lear (i=chatzill@rockbox/developer/lear)
20:09:39amiconndany_21a: Nope, the opposite
20:09:56dany_21aamiconn: damn... :)
20:11:52 Join Buschel [0] (
20:13:16*Buschel needs to remeasure svn mp3...
20:13:21Buschelthe results cannot be true
20:14:25*Buschel just measured the c-dct32 as 8.25MHz faster than the asm one
20:14:49 Join Acksaw [0] (
20:20:18 Join neddy [0] (n=john@nat/sun/x-668d9e283e1c638b)
20:21:19mcuelenaerethread.c has this line: "thread->context.sp = (typeof (thread->context.sp))stackend;" of what type does context.sp needs to be?
20:21:44mcuelenaere(GCC gives me "error: 'struct regs' has no member named '$29'")
20:21:54gevaertsI would guess the stack pointer
20:22:12mcuelenaereme too, but isn't that an unsigned int aka uint32_t ?
20:23:14 Quit Acksaw (Read error: 104 (Connection reset by peer))
20:23:26mcuelenaerestackend is a uintptr_t BTW
20:23:35gevaertsProbably, but that error seems to point to a different problem I think
20:23:55Buschelok, remeasurements looks reasonable fir libmad. c-dct32 -O2 = 39.75MHz, c-dct32 -01 = 38.27MHz, asm dct -O2 = 38.28MHz
20:24:11mcuelenaereI think the code will work with a cast to (uint32_t), but I'm not going to change that as it's not cross-platform compatible
20:24:19Buschelso, at least the c-dct is of the same speed as the asm'ed dct
20:25:05mcuelenaereah never mind, gevaerts you were right: the problem is elsewhere
20:27:48amiconnBuschel: Did you also test the PP5020 case? (less iram - dunno whether libmad on arm makes use of that)
20:28:07amiconnPP5002 is probably more different
20:28:53Buschelamiconn: no, just have a 5022 here. just trying to put the c-dct32 into iram now
20:29:25amiconnYou can test the 5020 case. PP5020 and PP5022 are identical in speed as long as the same things are in iram
20:29:48amiconnI mean you can test it on PP5022
20:30:04 Join Acksaw [0] (
20:30:07Buschelamiconn: c-dct32 -O1 with iram = 38.26MHz
20:31:32Buschelahem, this was for -O2, doing -O1 now
20:32:19 Join bertrik [0] (n=5a80bda4@gateway/web/cgi-irc/
20:34:46Buschelc-dct32 -O1 with iram = 38.25MHz
20:36:43wpyhhi Buschel
20:36:53amiconnSo... pretty minimal differences
20:36:57wpyhI've got some new results at the IpodRuntime page
20:37:15amiconnWhat's the margin of error in those measurements?
20:37:15*wpyh is not going to disturb their conversation
20:37:32Buschelamiconn: they are quite reproducible
20:37:45amiconnYeah, but +/- 0.02MHz ??
20:37:47Buschelamiconn: max difference is +/- 0.01MHz
20:38:03amiconnThat's less than one permille
20:38:55mcuelenaereare #define's included from a .h file in a .c file still valid when it's run through the assembler stage?
20:39:01Buschelamiconn: how can we build synth.c with -O1, but the other files with -O2?
20:39:09amiconnAnd alignment can probably change things more than that. On coldfire it definitely does; didn't see that effect on arm yet though
20:39:45amiconnBuschel: I'd just keep using the asm version....
20:40:07amiconnMaybe that one can be optimised further, and then you'll gain more than 0.02MHz
20:40:55Buschelat least I would learn more about makefiles ;-)
20:41:15*wpyh sees that the conversation has died down somewhat
20:41:25wpyhBuschel:from my data on the IpodRuntime page
20:41:42wpyh(there's a final data missing but it should follow the trend)
20:42:01Buschelwpyh: yes, i saw them. libmad does not scale that much with the bitrate as the OF does
20:42:15wpyhthe runtime increases that we see when we use a lower bitrate codec is lower than with the OF
20:42:19Buschel(or better: rockbox does not scale...)
20:42:32wpyhoh? I was thinking about the buffering algorithm, but I'm not sure
20:43:31wpyhand the OF doesn't turn off accessory power supply, right? while in my tests, I turned it off... so there must be a huge power eater somewhere
20:43:34amiconnBuschel: If you want that, you'd at least need to test on *all* arm variants
20:44:01 Nick amiconn is now known as PowerEater (n=jens@rockbox/developer/amiconn)
20:44:01mcuelenaereto clarify my question: there are some registers who need to be replaced by #define's, but those are in an asm volatile() statement; is it possible to do so?
20:44:09***Saving seen data "./dancer.seen"
20:44:10 Nick PowerEater is now known as amiconn (n=jens@rockbox/developer/amiconn)
20:44:28Buschelamiconn: i would like to know what the effect on _my_ player for now
20:44:29 Join spiorf [0] (
20:44:43 Quit spiorf (Remote closed the connection)
20:44:56wpyhbtw I will make further tests on my Nano −− it has a shorter battery runtime so tests will finish more quickly ;)
20:45:54*amiconn wonders how Buschel wants to measure runtime differences in the 10-seconds range...
20:48:05*Buschel is a bit penetrant regarding codec performance
20:48:32*Buschel is mixing up german and english :/
20:49:37 Join EspeonEefi [0] (i=espeonee@STRATTON-FOUR-EIGHTY-EIGHT.MIT.EDU)
20:50:50 Join hannesd [0] (
20:50:50bertrikBuschel: you use battery bench a lot, right? Can you have a look at this simplified version of battery bench that I posted here: ?
20:51:33Lloreanbluebrother: "Installation instructions can be found in the manual, which is (here). The recommended method is by use of RBUtil."
20:51:52GodEater_I think we should probably remove the bias in the instructions towards the "manual" method too
20:52:00GodEater_maybe move the manual method to an appendix even
20:52:00Nico_PLlorean: I was thinking of something along the lines of adding one line underneath "Why should you run Rockbox? Click here to find out" with something like "Installation instructions(link) - Read the manual(link) for more detailed instructions on how to use Rockbox"
20:52:05bluebrotherLlorean: yes, something like that.
20:52:40 Join Mathiasdm2 [0] (
20:52:55*Nico_P agrees that rbutil needs to be more visible
20:53:00bluebrotherI also had the idea of adding a startup screen to rbutil (mainly for checking system stuff and giving a warning about issues immediately). We could add a hint to the manual during that screen as well
20:53:33GodEater_bluebrother: how many targets have you had connected to your PC at once with rbutil running btw ?
20:53:37bertrikin this version of battery bench, the instant that a measurement is done no longer depends on disk activity, disk spin-down times or whether music is playing or not, it simply takes a measurement each minute
20:53:48GodEater_I noticed the other day it gets easily confused with multiple DAPS attached at once
20:53:54bluebrotherGodEater_: good question. Let me try with as much targets as available ;-)
20:54:01GodEater_I've done it with 5 attached
20:54:17bluebrotherany major problems with that?
20:54:24LloreanActually, 1.0.5 gets confused if you have a .rockbox folder on a local HD
20:54:28domonokythe detection wont really work nice with multiple devices...
20:54:30LloreanSay you extracted a build there to tinker with the files of.
20:54:39LloreanIt detected my "D:" drive as an iPod nano
20:54:59domonoky:-) which is very nice for testing :-)
20:55:03bluebrotherif you faked your drive D: as being a nano ... ;-)
20:55:09LloreanI'd say moving "Manual Install" to an appendix is a very good idea.
20:56:07GodEater_with my setup (linux version) it misidentified quite a few of the targets
20:56:08domonokyits the detection via the file, its always first, and just uses the first drive with this file found :-)
20:56:09 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
20:56:13Buschelbetrik: hmm, i should run a new battery bench with this patch then
20:56:45bluebrotherdomonoky: not exactly. USB detection comes first ;-)
20:57:05amiconnbertrik: Btw, JdGordon's remark is incorrect. The ata callback on flash targets is called after normal disk activity like on hdd targets
20:57:13domonokybluebrother: yes, but usb never stops the detection, so the stuff overrides..
20:57:15*mcuelenaere really hates this MIPSel assembler..
20:57:28*GodEater_ doesn't care if people simple give up, it's the ones who don't give up, and come and ask for advice afterwards that are annoying
20:57:32bluebrotherhmm, right. We should change that ...
20:57:50amiconn(see e.g. ata-sd-pp.c and ata_mmc.c)
20:57:55LloreanGodEater_: This is in response to?
20:57:56bertrikamiconn: ok thanks, I hadn't really looked into it myself yet
20:58:03domonokymaybe we should always try to detect everything, and present a list, if there are multiple devices ...
20:58:12bluebrotheror at least try to make detection more robust. But I'm still hoping to add some other missing mountpoint detection first
20:58:14GodEater_Llorean: your forum post ;)
20:58:32GodEater_I thought it was too cheeky to add to the actual thread
20:58:35 Join stoffel [0] (
20:58:49*GodEater_ again wishes for meta-comments :)
20:59:39domonokybluebrother: we should also use all usb ids we can use, and only use the patchers as a last resort (because they need special permissions)...
20:59:41LloreanGodEater_: I clarified it a bit.
21:00:06bluebrotherdomonoky: well, the problem is that some Ipods have the same PID
21:00:09LloreanI see nothing wrong with not caring if people give up, especially if they might increase the load on our support people for no real gain. So I don't see any reason to encourage them *not* to read the docs.
21:00:19 Join obo [0] (n=obo@rockbox/developer/obo)
21:00:32bertrikBuschel: yes please test it if you can. I am not able to test it on other targets than sansa e200 myself.
21:00:41GodEater_I'm not sure we have a way of measuring what percentage give up entirely, and what percentage give up on their own and then come and bug us
21:00:44domonokybluebrother: yes, but we only have to use ipodpatcher if we detect one of the dublicates..
21:00:48bluebrotherok, this is bad: I now have 5 players connected (of which one is a fake Ipod). Rbutil detected one of the players but used a wrong mountpoint
21:01:06GodEater_bluebrother: that's exactly what I found
21:01:14bluebrotherdomonoky: in that case it would be good if we could make rbutil ask for elevation on vista not earlier than needed.
21:01:32 Quit neddy ("Leaving.")
21:01:40LloreanGodEater_: I'd say that simply putting an installation doc together, with the installer linked inside it, rather than suggesting they start with the installer and try to stumble though, is better.
21:01:41domonokybluebrother: would be nice, if i only would know how to this..
21:01:45bluebrotherI think the easiest soltion would be to simply not set the device / mountpoint and present an error if multiple devices are detected.
21:01:48GodEater_bluebrother: I'm not saying it's going to be a common occurence, I just thought you should know about it
21:01:50LloreanIf you ask them to stumble through, I think they're MORE likely to come ask for help without reading anything, not less.
21:01:54bluebrotheryou're the guy working on windows ... ;-)
21:02:16GodEater_Llorean: absolutely - couldn't agree more
21:02:34bluebrotherGodEater_: I agree that we should do something about that. Don't consider it problematic or anything above lowest priority though ;-)
21:02:50GodEater_no - it's only us mad fools with multiple targets generally :)
21:02:57bluebrotherhehe :)
21:02:58GodEater_and I don't often update them all at once
21:03:28GodEater_it's definitely an "edge case"
21:04:51bluebrothera short "installation with rbutil and screenshots" might be nice.
21:05:08Lloreanbluebrother: Agreed
21:05:09 Quit delorean90 ()
21:05:27 Nick fxb__ is now known as fxb (
21:05:34LloreanI'd like to see an installation instructions that was target independent, since with the exception of one or two special steps (that RBUtil should tell you clearly about, in the ideal case) it can be.
21:06:03bluebrotherI might try putting one up in the wiki so we can improve it and eventually create some manual-like installation document from it
21:06:37 Quit Mathiasdm (Read error: 110 (Connection timed out))
21:06:41LloreanSounds good to me.
21:07:26LloreanHis work life seems to have picked up around a similar time it happened.
21:07:58LloreanBut the "Mod" position is more in dealing with bans, problem users, etc.
21:08:23LloreanSince I know he's more moderate than me, and I'd met him at DevConW, etc, I'd decided he seemed like a good role for the job, and felt we could probably use one more mod
21:08:49GodEater_is there anyone *less* moderate than you? :)
21:09:11LloreanI think I banned 'em all.
21:10:16 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
21:10:17LloreanAnyway, I'm thinking of having one more "Global Mod" for the forums sometime soon. They're not only getting big, but probably with a rules reorganization and all the recent upcoming ports, they'll get a lot more active.
21:10:21 Quit pondlife ("Leaving.")
21:10:32GodEater_very true
21:10:38GodEater_the D2 looks close
21:10:40LloreanAnd I hate the fact that it's true, but more activity means we need a few more people with hammers.
21:10:42GodEater_the GBS is *very* close
21:10:59LloreanI think we've already had the brunt of the GBS rush dimmed by how long it's been in that nether area it's in.
21:11:13GodEater_I'm skeptical of that
21:11:21GodEater_I think what it goes on the front page it'll pick up again
21:11:33LloreanI think it'll be a good deal less than it would've been if it happened two months ago, though
21:11:33GodEater_although not hugely perhaps, given how hard it is to come by
21:11:46LloreanThe SansaV2 is where I think the biggest rush will come in.
21:11:57GodEater_that's going to be "fun"
21:12:26*GodEater_ remembers "fondly" the last influx of sansa users
21:12:39LloreanThat's a big part of why I think the forums need simpler rules, and consistent enforcing
21:12:41GodEater_though most of them have mellowed with time ;)
21:12:43bluebrotherLlorean: what do you think about "GraphicalInstall" as pagename?
21:12:46LloreanWe need something that scales well.
21:13:00Lloreanbluebrother: Works for me, and I think it'll help make people feel comfortable.
21:13:58*gevaerts was probably one of the latecomers in this last influx of sansa users
21:14:12Lloreangevaerts: You certainly aren't in the group we're referring to.
21:14:29GodEater_hell no
21:14:36LloreanThough it's possible you've "bricked" your player as often as they have...
21:14:46GodEater_unless I forgot you drooling into your beard when you first arrived
21:14:54GodEater_and smacking your knuckles on the door step
21:15:10LambdaCalculus37Hehehe... :)
21:15:21GodEater_I'm pretty sure I'd remember though
21:15:33gevaertsI'm probably the only one to have "bricked" his c200 to be able to give exact instructions to someone else :)
21:15:44LambdaCalculus37Now that's dedication. :)
21:15:46GodEater_what, on how to brick it exactly the same ? :)
21:16:13GodEater_1) Take one microwave oven....
21:16:21GodEater_2) insert your DAP...
21:16:24gevaertsThat's for bricking, not "bricking"
21:16:39GodEater_ah, that subtle difference some many of our users miss
21:16:45LambdaCalculus373) Microwave on HIGH for 5 minutes.
21:16:54GodEater_4) Stand well back
21:17:14LambdaCalculus375) Boom!
21:17:17LambdaCalculus376) Hahaha!
21:17:40gevaertsOr just 2) drop microwave (with DAP in it) in a handy nearby river
21:18:15LloreanI'm also not entirely confident about who we'll be getting with the Meizu player
21:18:38GodEater_they're going to be hard work
21:18:40LambdaCalculus37Llorean: I can pretty much assure that we're going to have a massive influx of visitors from China with the Meizu.
21:18:49LloreanI think the big fear for me is always when I go check out their own forums, and realize they're going to come to ours and expect to be able to behave the same
21:19:08*GodEater_ gets ready with the big stick to wave at them
21:19:44LambdaCalculus37Llorean: I had a crazy idea... why not also assign specific people to mod various areas of the forums? For example, someone can mod the Apple section, another mod the Toshiba section, etc...
21:19:53 Quit OlivierBorowski_ (Remote closed the connection)
21:19:57LloreanLambdaCalculus37: I've actually been intentionally trying to avoid that.
21:20:27LambdaCalculus37Llorean: Understood. Sorry for bringing it up. :)
21:20:27LloreanI think that just means I'm telling people exactly who to pester if they have a question about TopicX and don't want to post it in a thread
21:20:33 Join OlivierBorowski_ [0] (
21:20:58LloreanThe forums have functionality specifically for it, but really, if I'm going to give someone full "moderator" type privileges, it's probably just going to be globally
21:21:30LambdaCalculus37Actually, that does sound like the better idea.
21:21:43GodEater_LambdaCalculus37: how much do you consciously spend avoiding bits of the forum in your own browsing habits ?
21:21:46LloreanI mean, basically, making someone a mod over a specific area is saying "I trust you to moderate, but not to know when you don't know enough to moderate"
21:22:07 Quit Buschel ()
21:23:04LambdaCalculus37GodEater_: I don't go into the Olympus section, and I don't get involved with the unsupported build section, and it irks me to see questions about unsupported builds being asked in the wrong areas.
21:23:59Lloreanthe Olympus section hasn't had a new post in forever.
21:24:02LloreanIt's not the most lively target
21:24:03GodEater_LambdaCalculus37: you should do what I did
21:24:07LloreanEither that, or it has the brightest users _ever_
21:24:12GodEater_take a 2 month sabbatical
21:24:19GodEater_I'm *so* chilled about the forums right now :)
21:24:39GodEater_I've not been snide to anyone for weeks
21:25:17 Quit bertrik ("CGI:IRC (EOF)")
21:27:37LloreanWell, ideally the pre-written responses for each of the possible rules violations will mean a vastly overall reduced "snideness" anyway. Unless people decide the canned responses are snide.
21:27:38MarcGuayIt's unfortunate that there's no voicing of "recording started"/"recording stopped". I can only imagine how difficult it would be to know what's going on otherwise.
21:27:42 Join billiob [0] (
21:29:25 Join neddy [0] (n=john@nat/sun/x-2861224ba0b8c2da)
21:29:44LloreanMarcGuay: I'm not certain, but isn't there a way to beep on stop/start?
21:30:27*MarcGuay goes to find out
21:31:18MarcGuayDon't see anything on the e200.
21:31:47LloreanI think it might be the normal button beep
21:32:49GodEater_do you mean key click ?
21:33:08MarcGuayLlorean: Keyclick + Start recording just locked up my player... :)
21:33:22LloreanWell, obviously that's not ideal behaviour
21:33:22 Join Ors [0] (
21:33:27 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
21:33:28GodEater_obviously this is intended behaviour ;)
21:33:32MarcGuayFBut lyspray isn't there.
21:33:59MarcGuayClose enough.
21:35:03MarcGuayCrud. I can reproduce that lockup. So much for replying to that guy on the list..
21:35:47mcuelenaereis memset16 equal to memset with the difference that the first one is optimized for word access?
21:35:48MarcGuayKeyclick seems to be unreliable with the voices menus in general, sometimes beeping, sometimes not.
21:36:19 Quit Arathis2 ("Bye, bye")
21:36:25GodEater_MarcGuay: imo Keyclick is bloody dreadful across all PP targets
21:36:43MarcGuayGodEater: It's supposed to be a beep or a click?
21:36:49GodEater_heck if I know
21:36:53LambdaCalculus37Agreed. I can hardly use it on either one of my iPods.
21:36:57pixelmaLlorean, MarcGuay: I think I remember the Archoses which can record can "beep"
21:36:59GodEater_it made my ears bleed the first time I tried it
21:37:08GodEater_I've never turned it on since
21:37:26LloreanSo, what you're saying is, "keyclick should be ifdeffed out for the release"?
21:37:37LloreanAt least for PP targets?
21:37:40pixelmaGodEater: pondlife fixed that but it means that it doesn't work 100% of the time...
21:37:42GodEater_well I would
21:37:54LloreanGodEater_: Mind posting a bug report on it (what it does, how to reproduce if possible, etc)?
21:37:54GodEater_pixelma: so it's still somewhat useless then ?
21:38:11GodEater_Llorean: well if pondlife "fixed" the ears bleeding thing, there's not much point
21:38:17GodEater_I think I documented that on the original patch anyway
21:38:22GodEater_he just ignored it ;)
21:38:29pixelmaand that beeeep only occured to me (on the c200) when voice was enabled), if it was disabled it worked correctly
21:38:33*Llorean needs a list of "features that don't work, or do work but do something stupid"
21:38:40pixelmanot sure what it does currently haven
21:38:43MarcGuayLlorean: Yeah, it's awful. I just had it pierce my ears (even on weak). It froze on "meeeeeep" until I restarted playback.
21:38:45*GodEater_ thinks Llorean will get a long list
21:38:55pixelmahaven't tried for a while
21:39:06LloreanGodEater_: I WANT a long list.
21:39:09Lloreanthe longer the better.
21:39:20gevaertsLlorean: is that the thing called "release notes" ?
21:39:23LloreanThen we can decide "which ones do we remove, and which do we put in the Known Issues file"
21:39:26LloreanWho cares about fixing things.
21:39:26 Join hannesd_ [0] (
21:39:29MarcGuayThe long meep was with voicing off as well.
21:39:38Lloreangevaerts: Exactly!
21:39:47OrsHi there
21:40:57 Join webguest93 [0] (n=4e0c81cb@gateway/web/cgi-irc/
21:41:06OrsCould somebody help me and tell whether my Zicplay mp3 player can be used with rockbox? I could not find it in the listing
21:41:24LloreanOrs: It's not in the listing because it cannot be used.
21:41:31LambdaCalculus37Only the players on the front page can be used.
21:42:03 Quit OlivierBorowski_ (Remote closed the connection)
21:42:25OrsOK. I just asked to be on the safe side, any other way I could use my mp3 player, alternative solution?
21:42:30MarcGuay <- Recording hardlocks with keyclick enabled
21:42:49MarcGuayMark it as 3.0?
21:43:17*Llorean goes to do that
21:44:35LloreanMarcGuay: If you can mark things as 3.0, feel free to mark anything at all you think even *might* be 3.0 relevant. I think it's better to have a large list to pare down, rather than later wondering if we need to go back through the non 3.0 stuff again and again.
21:44:41LambdaCalculus37Ors: Buy a compatible player?
21:44:54MarcGuayLlorean: I agree.
21:45:38LloreanGood. :)
21:45:55 Quit webguest93 ("CGI:IRC (Ping timeout)")
21:45:55OrsLambdaCalculus37: I have meant to use this old player
21:46:13GodEater_Ors: in that case your only option is "port rockbox to it yourself"
21:47:24 Quit BigBambi (Read error: 113 (No route to host))
21:50:42OrsI guess that would be rather complicated for me...
21:50:42 Quit stoffel ("Lost terminal")
21:51:34 Quit arohtar (Client Quit)
21:52:05 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
21:52:14 Join PaulJam_ [0] (
21:52:37LambdaCalculus37Ors: Well, then you're not getting Rockbox.
21:53:37 Join remghoost [0] (
21:53:40Orsso I thought :-(
21:53:41gevaertslinuxstb: am I right that the current dax status basically is "code runs, buttons are known, and lcd works" ?
21:54:06linuxstbMore or less. Interrupts are also working, so the tick task is running, and sleep works.
21:54:25shotofaddsI think the current status is waiting for "someone" to play with the NAND driver
21:54:42remghoosthello everyone
21:55:03*shotofadds has LambdaCalculus37's m240 now, so might take a look soon
21:55:10gevaertsMine was switched to usb boot today, so I can play with it. I'm planning to take it with me on holiday
21:55:18 Quit hannesd (Read error: 110 (Connection timed out))
21:55:18 Nick hannesd_ is now known as hannesd (
21:55:31bluebrotherok, up for comments and improvements:
21:56:15remghoostso, the rockbox installer says that i have an unsupported apple player variant. what should i do?
21:56:51linuxstbremghoost: Either buy a new player, or follow the step-by-step guide here ;)
21:57:13 Quit LambdaCalculus37 (" ajax IRC Client")
21:57:16remghoostalright, thanks so much
21:58:06gevaertsbluebrother: looks good to me
21:58:40bluebrotherI hope those windows kids can cope with the screenshots made on linux ;-)
21:59:23gevaertsThey can, but what about the OSX kids?
21:59:42bluebrotherwell, those are such a minority ... ;-)
22:00:04 Quit remghoost ()
22:01:03bluebrotherreally nice that rbutil now detects unsupported Ipods.
22:01:59gevaertsIndeed. You can consider that tested now :)
22:02:16GodEater_Llorean: you realise "he" is the original poster in the ipod skipping thread ?
22:02:23bluebrotherI did test it. I made my pendrive being an unsupported Ipod :)
22:02:55amiconnAh, linuxstb da breakah is back
22:03:23LloreanGodEater_: Oops
22:03:27GodEater_when did amiconn become a gansta rapper ?
22:03:31LloreanI read "same problem" and assumed different person.
22:03:38LloreanOtherwise I'd have expected "still have the problem"
22:03:47*GodEater_ is here to help Llorean avoid looking silly
22:05:27 Quit PaulJam (Read error: 110 (Connection timed out))
22:06:20 Join spiorf [0] (
22:09:07 Part Ors ("Leaving.")
22:12:26 Quit spiorf (Remote closed the connection)
22:12:50 Join spiorf [0] (
22:14:22gevaertsbluebrother: maybe the wording on that first dialog can be a bit better. Would it be hard to distinguish between "no configuration" and "invalid/wrong configuration" and give a friendly "welcome to rockbox, you need to setup some things" message in the first case?
22:14:41*gevaerts also asks domonoky
22:15:16domonokygevaerts: sure.. feel free to modify.. :-)
22:16:38 Join stripwax [0] (
22:16:50shotofaddsGodEater_: do we have a concept of "supported" codecs now?
22:16:59*Nico_P was thinking the same as gevaerts
22:17:09*gevaerts is looking at the source
22:18:21shotofaddsgevaerts: while you're there, a minor nit-pick: "allow you to correcting" should be "allow you to correct"
22:18:35GodEater_shotofadds: not really, but we have ones which work better than others
22:18:56shotofaddsyeah, it's just that s-word again...
22:19:47GodEater_feel free to change my wording ;)
22:20:06gevaertsshotofadds: I'm on it :)
22:20:27GodEater_shotofadds: that better ?
22:20:47shotofaddsspot on
22:21:00Nico_Pbluebrother: nice guide :) I'm not sure WikiManual is the best parent for it though
22:23:48LearNico_P: Looked at the patch yet? (FS #9164)
22:24:36 Quit midgey ()
22:24:51Nico_PLear: yes, but I'm not really in shape to think serioulsy about which mutex locks are really necessary
22:25:16MarcGuayAnyone feel like testing a new keymap for the iPod recording screen?
22:25:19Nico_PLear: have you tried to isolate the ones that actually fix your issue?
22:26:24 Join midgey [0] (
22:28:35*gevaerts has fixed the easy bit : the correcting thing. Now the hard bit...
22:29:22 Quit midgey (Read error: 104 (Connection reset by peer))
22:30:04LearNico_P: Not really. I do know that the one in bufopen wasn't enough though. Maybe not required either. :)
22:30:08gevaertsbluebrother, domonoky: would adding a boolean field to indicate whether the current config comes from a file (set to true on sync()) to rbsettings be a good way to solve this ?
22:30:08bluebrothergevaerts: I think some source texts could get made better. Starting from obvious errors ("plenken") up to no-really-user-compatible wording
22:30:53bluebrothergevaerts: I think the better solution is to add a new "welcome" screen.
22:31:19gevaertsYes, but that still needs no-existing-config detection
22:31:21bluebrotherthat would include a "don't remind me again" button which is such a flag. It could also inform the user if permissions are missing or similar
22:31:25Nico_PLear: yeah I'm not sure the one in bufopen is required
22:31:43 Quit Mathiasdm2 ("Invisible Internet Project:")
22:31:52bluebrotheralso, I wanted to make this screen re-appear on each update so we can present the users a changelog
22:32:15*gevaerts is starting to think that this may not be a five minute fix
22:32:38bluebrotheryep, that's the reason I haven't done it
22:32:47bluebrotherbut its on my todo list
22:33:17gevaertsThen I'd probably better not do a quick fix now, to avoid interfering with whatever you come up with
22:33:38gevaertsIt's just a bit of ugliness after all, not a critical bug
22:34:06 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
22:34:14Nico_PLear: add_handle already handles the mutex, and there's nothing that yields in bufopen apart from what was already protected
22:34:27Nico_Pso I don't think the changes there are needed
22:34:47Nico_Palso, what about the last hunk?
22:35:06LearNico_P: I added that mainly because add_handle and load_bitmap both mess around with various "globals". And mutex_lock can yield, can't it?
22:35:14domonokygevaerts: you could start with a "bool firstStart()" function in the setting object, and show a nicer Dialog on first Start...
22:35:21bluebrothergevaerts: feel free to open a task in the tracker as a reminder ;-)
22:35:24domonokythis Diaglog can the be improved later.. :-)
22:35:37bluebrotherdomonoky: I want to save the current version info instead ...
22:36:22 Part billiob
22:36:47Nico_PLear: no, it doesn't yield. even is they both touch the same vars, it's not really a problem as long as they do it sequentially, which is already guaranteed
22:37:29gevaertsSo maybe a QString curVersion() function, and a new dialog if that is nonexistent or different from the software version ?
22:37:36LearNico_P: shrink_buffer walks the handle list and moves them around.
22:39:36 Quit fyrestorm (Read error: 113 (No route to host))
22:39:45Nico_PLear: the moving is already protected by the mutex, inside (move_handle)
22:39:57domonokygevaerts: for me this sounds good... lets see if bluebrother has other plans :-)
22:40:55bluebrotherdomonoky: you know I'm not happy with the settings object anyway
22:41:16*gevaerts doesn't volunteer to rewrite _that_
22:41:16domonokyyes, it could need some improvement...
22:41:32*bluebrother would do it but still has the bootloader class to finish
22:42:14Nico_PLear: actually, correct me if I'm wrong, but I'm starting to think that maybe none of those additional mutexes are needed and that the only issue was m being made out of date by shrink_handle
22:42:28domonokybut it doesnt hurt to add another function to it, till we find time to redesing it.. :-)
22:44:05*gevaerts goes ahead, and hopes to have a patch ready to review in a few decaminutes
22:44:13***Saving seen data "./dancer.seen"
22:48:18LearNico_P: Possibly. On the other hand, if the albumart bufopen was done too quickly, I got the abort - while doing the initial buffering. Would shrink_handle really do anything in that case?
22:52:23Nico_Pmaybe it moves the audio handle, yes
22:54:31 Join fyrestorm [0] (
22:55:52Nico_PLear: would you mind trying without the added mutex calls?
22:55:58 Quit atrus (Read error: 110 (Connection timed out))
22:56:15Nico_PI'm curious to know whether it would still work, although I have doubts
22:56:23LearI'm working on it. Need to reproduce it first... :)
22:56:48 Quit hannesd (Read error: 148 (No route to host))
22:56:53Nico_Pall right :)
22:58:25 Join culture [0] (
22:59:33 Join delorean90 [0] (
22:59:57gevaertsbluebrother, domonoky: where is the configuration file saved ?
23:01:17 Join bertrik [0] (n=5a80bda4@gateway/web/cgi-irc/
23:01:17 Quit jgarvey ("Leaving")
23:01:21domonokydepends on the plattform...
23:01:28*gevaerts found it
23:01:40gevaertsI just wanted to check if my new string is in it
23:01:47*bluebrother simply assumed linux :)
23:04:55domonokythis link shows where its store on the different plattforms (we use ini format)
23:09:18Nico_PLear: btw, thanks for looking into the code... it's always good to have some fresh eyes :)
23:14:04LearNico_P: You're welcome. But now I can't reproduce it. Sometimes it is easy, sometimes not...
23:15:19 Part domonoky
23:21:36 Join PaulJam [0] (
23:25:49 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
23:28:00*gevaerts wants to know why this thing doesn't want to work
23:29:33gevaertsbluebrother: if I do a settings->setWhatever(); just before a settings->sync(); shouldn't the new setting be saved?
23:29:50 Part dany_21a
23:30:31*gevaerts has a suspicion of what might be wrong
23:35:42gevaertsbluebrother: around?
23:36:19 Quit PaulJam_ (Read error: 113 (No route to host))
23:37:30gevaertsbluebrother: patch at Do you want me to commit it, or just open an FS task?
23:38:51 Join goffa [0] (n=goffa@
23:39:04*gevaerts has been searching for at least twenty minutes why two strings were different. Turns out that for some strange reason __TIME__ changes everytime you build, and can be different for different files
23:39:09 Quit bertrik ("CGI:IRC")
23:39:09bluebrothergevaerts: sorry, was distracted for a bit
23:40:01bluebrotherlooks nice. Go for it!
23:40:09gevaertsbluebrother: no problem. I lost more time to being stupid than to your distraction :)
23:40:24 Quit xqtftqx (" ajax IRC Client")
23:41:26goffahmm... may have found a bug on my gigabeat, but i'm having trouble isolating it.. i use the random folder advance, when i advance to the next folder using short right/long right it goes to the next folder but i get no audio
23:41:35goffathe song plays, but silently
23:41:42goffaif i restart the track using back, it plays
23:41:53goffadoesn't do that every time
23:42:47goffaalso... when i do short left long left it doesn't go back to the previous dir.. just to a random one
23:42:53goffadon't know if its supposed to
23:44:06 Join stripwax [0] (
23:44:10goffaguess that's two bugs.. but i don't know if the second one is a bug :)
23:44:11MarcGuaygoffa: If it's random, I'd expect it to be random in both directions, I guess.
23:44:47PaulJamgoffa: for the first issue see FS #9152, the second one is intended.
23:45:17goffaalthough.. random backwards should go back imo
23:45:23 Join stripwax_ [0] (
23:45:27 Quit stripwax (Read error: 104 (Connection reset by peer))
23:46:02goffabut might be too much code on the other hand
23:46:25linuxstbSo what's the difference between random forwards and random backwards?
23:46:56PaulJamthere is no difference (in the current implementation)
23:46:57goffawell... backwards shouldn't be random
23:47:23linuxstbSo what's the point in the two button mappings? Why not just use one?
23:47:52PaulJamlinuxstb: for non-random dirskip
23:48:19linuxstbIMO, skipping backwards should be disabled in random mode then - or fixed.
23:49:03Nico_Pgoffa: play/pause will "fix" the no sound issue. could you add a comment to the task PaulJam linked you to with a description of how you got it?
23:49:15MarcGuaygoffa: Does the same thing happen if you use the combo-keypress?
23:49:34goffahmm... i never do that.. let me try a couple times
23:49:52MarcGuaygoffa: Curious about both issues...
23:50:40goffajust did it about 20 times
23:50:42PaulJamNico_P: when you got the issue, did you use the PLAY + LEFT/RIGHT buttoncombo or the short - long combo?
23:50:49goffano error
23:51:12Nico_PPaulJam: actually I haven't gotten it that way for a long time because I never skip dirs
23:51:12goffaso it would seem that its just short right long right
23:51:19goffai didn't try combo left
23:51:25goffamostly because i don't care :)
23:51:26MarcGuaygoffa: And does it random backwards as well?
23:51:32goffai can do it
23:52:57goffaok.. combo left seems to work too
23:53:11goffalet me try short left long left
23:53:16MarcGuaygoffa: It goes to the previous album non-randomly?
23:53:37 Join stoffel [0] (
23:54:03goffaMarcGuay: it goes to a random album
23:54:13goffahowever.. there seems to be consistant sound
23:54:43MarcGuaygoffa: The difference between the two seems to be in gwps.c starting at line 390.
23:54:44goffai hope that made sense
23:57:54MarcGuayLine 363, actually, for the skip fwd.
23:58:28goffahere's the way i think it should behave... maybe i'm illogical... left: beginning of track, double left: last track of previous album. combo or short long first track of previous album
23:58:45 Join rocko_ [0] (

Previous day | Next day