Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2008-05-11

00:02:53 Join einhirn_ [0] (
00:03:59jhMikeSkkurbjun: flash away
00:04:35jhMikeSI had already done that for the S bootloader but neglected to just finish the job :p
00:05:37peturmcuelenaere: thanks for fixing my typo ;)
00:07:38XqtftqxHey Guys, Just want to say, the gigabeat s port is realy moving along, and everybody had bean doing a great job, im running it currently. Great Stuff!
00:07:52mcuelenaerepetur: np ;)
00:09:00XqtftqxAnybody Know the status of the battery charging thing for the S?
00:09:29jhMikeSXqtftqx: yes, in the planning stage
00:09:42XqtftqxWhats the status
00:09:44 Quit yeal` ()
00:10:02jhMikeSexactly that :)
00:10:13XqtftqxWell that realy helps
00:10:19 Quit mcuelenaere (""Zzzz..."")
00:10:24Xqtftqxare you guys stuck on something, i can try to help out?
00:11:12 Quit ajkessel ("BitchX: now with 42 percent more random quit messages!")
00:11:16 Join Xqtftqxt [0] (i=40e9e3d2@gateway/web/ajax/
00:11:20jhMikeSnot stuck, just thinking it through before twiddling fingers
00:11:25kkurbjunjhMikeS: looks good, just need to fix the bootloader link script
00:11:37kkurbjunflash is still working
00:11:51XqtftqxtWhat are you guys talking about?
00:11:58 Quit Xqtftqx (" ajax IRC Client")
00:12:24 Nick Xqtftqxt is now known as Xqtftqx (i=40e9e3d2@gateway/web/ajax/
00:12:29XqtftqxFirefox Crashed
00:13:05kkurbjunjhMikeS: I just committed the fix
00:13:12Xqtftqxwhat fix?
00:13:25LloreanXqtftqx: Please, read the logs if you want to know what's being talked about.
00:13:34XqtftqxOk, ill read, sorry
00:13:50XqtftqxI cant
00:13:54Xqtftqxonly goes back so far
00:14:10XqtftqxOk thanks
00:14:19LloreanXqtftqx: Please, read the channel guidelines too. Try to use complete sentences rather than posting several very short lines.
00:14:37kkurbjunXqtftqx: I was talking about the F/X bootloader, not the S
00:15:11XqtftqxOh, Alright
00:15:46 Quit blo0dy ("changing servers")
00:16:02 Join blo0dy [0] (
00:17:13 Join simonrvn [0] (i=simon@unaffiliated/simonrvn)
00:18:06jhMikeSkkurbjun: thanks
00:18:34*jhMikeS spotted-up the nice, green build table today :\
00:20:32 Nick fxb__ is now known as fxb (
00:22:53 Quit desowin ("KVIrc 4.0.0 Insomnia")
00:23:23 Join PrincessCaitlai [0] (
00:23:28 Quit petur ("Zzzz")
00:23:50 Part PrincessCaitlai
00:25:03 Join TKFeck [0] (
00:25:10TKFeckcsc` indeed
00:25:13 Part TKFeck
00:26:49 Quit davina (Remote closed the connection)
00:28:21 Quit domonoky (Read error: 104 (Connection reset by peer))
00:34:50 Nick Bright-SUSaiyan is now known as SUSaiyan (
00:39:12saratogajhMikeS: does this mean hacks are no longer necessary for the Gigabeats when putting code in IRAM?
00:41:00 Nick csc` is now known as turks (n=csc@archlinux/user/csc)
00:41:30 Join bertrik [0] (
00:41:59jhMikeSsaratoga: not now
00:42:28 Nick turks is now known as csc` (n=csc@archlinux/user/csc)
00:50:07 Join YoungProGuru [0] (
00:52:07 Join tholdar [0] (
00:52:47 Quit Synergy6 (Read error: 110 (Connection timed out))
00:54:34 Quit tvelocity ("Αποχώρησε")
00:57:12 Quit simonrvn (Remote closed the connection)
01:04:00 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
01:04:04 Join t [0] (i=tom@freenode/staff/tomaw)
01:05:10tholdarpoor t
01:06:02 Part t
01:06:02 Join t [0] (i=tom@freenode/staff/tomaw)
01:11:55soapAdam Kessel?
01:13:08soapWhat is Adam Kessel's IRC nick?
01:13:16tholdarnick cave
01:13:32 Join Thundercloud [0] (
01:14:15 Quit tholdar (K-lined)
01:15:12 Quit XavierGr (Nick collision from services.)
01:15:23 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
01:16:22 Quit n1s ()
01:17:23 Join kugel [0] (n=kugel@unaffiliated/kugel)
01:18:15kugelbuilding rockbox doesn't do well with make -j
01:18:34 Join mf0102 [0] (n=michi@
01:20:35kugel"make -j" is ok, but "make -j 2" is not ok
01:20:43soapreally? I normally make -j3
01:20:59soapajkessel is the man I want to talk to.
01:21:38kugelsoap: Don't you need a space between -j and the number?
01:21:54kugeli did -j2 some time until I noticed it does nothing
01:22:01kugelsome time ago*
01:22:31*soap doesn't use a space
01:23:05kugelwell, I noticed a major speedup when i started to use the space
01:23:28soapgive me three and I'll test it.
01:23:47 Quit gevaerts ("bedtime")
01:24:10kugelsoap: No, you seem to be right, since -j2 doesn't do well too
01:25:13kugelthe problem with -j 2 on my machine is, that this makes the tools (like bmp2rb) be built after they are needed
01:25:39kugelthis applies to a fresh svn only ofc, when those tools aren't compiled yet
01:28:46soapmake -j 2 should be the same as make -j
01:28:59kugelI expected that too
01:29:01 Quit ender` (" Consultation, n. Medical term meaning "to share the wealth."")
01:29:01soapmake -j 20 should be the same as make -j IIUC
01:29:36kugelmake -j means make -j [infinite] according to the man page
01:29:39soapwhere it automatically attempt to pick the correct number of threads.
01:30:17kugelAllow N jobs at once; infinite jobs with no arg.
01:31:34 Part kakazza
01:34:26 Quit csc` (Remote closed the connection)
01:35:38saratogakugel: you may need to rerun configure or something like that
01:35:50saratogait sounds like your makefile is messed up
01:36:37obokugel: I just had the same issue with bmp2rb after a make veryclean
01:37:15soapFWIW, on my computer...svn up, make && make clean (to fill ccache), and the timings after that: j3=50sec, j2=62sec, j=50sec, no j=96sec
01:37:31obosolved it by changing the Makefile produced by tools/configure, to ensure that tools is built before features
01:39:03***Saving seen data "./dancer.seen"
01:39:43 Quit bertrik ("bye")
01:41:41soapajkessel doesn't appear to be a member of the forums, at least not from his IRC client's IP address range.
01:42:10soapI really would like to talk to him regarding his edit of the IpodAccessories web page.
01:43:46soapwould a wiki admin be able to see what IP address he edited the wiki page from? If so perhaps we could discover his forum account.
01:49:00 Join simonrvn [0] (i=simon@unaffiliated/simonrvn)
01:52:52 Quit simonrvn (Remote closed the connection)
01:53:57kugelsaratoga: after bmp2rb is built, everthing runs fine. I said, it only happens on a very fresh svn
01:54:15saratogakugel: i don't have that problem, so i think its something wrong at your end
01:54:29 Quit Xqtftqx (" ajax IRC Client")
01:55:17kugelsaratoga: Have you tried a freshly checked out source or one where you have been compiling within at least once? As you see, obo know what I mean
01:55:42obokugel: I just commited a change to solve it - svn up
01:56:00kugelobo: Ok, that was fast :)
01:56:10*kugel also didn't know obo is a dev :/
01:56:29*obo has been far to distracted by real life recently
01:58:20soapif I were rich i'd make a SoapSummerOfCode
01:59:49 Quit gregzx ("ChatZilla 0.9.82 [Firefox]")
02:00:16 Join simonrvn [0] (i=simon@unaffiliated/simonrvn)
02:02:42kugelobo: Thanks works fine :)
02:02:59kugelsaratoga: The, my end wasn't wrong :P
02:06:20 Join ReKleSS [0] (n=ReKleSS@
02:07:36 Quit kushal_one ("Leaving")
02:07:49 Join BlakeJohnson86 [0] (
02:09:56kugelI think rockbox should ship some kind of exampe-music file, maybe even with a cover art
02:10:06kugelat least the source for having it in the sim
02:10:15 Quit YoungProGuru ("Ex-Chat")
02:17:50 Quit dabujo ("( :: NoNameScript 4.2 :: )")
02:18:53 Quit Thundercloud (Remote closed the connection)
02:28:57 Quit PaulJam (".")
02:29:53 Join tvelocity [0] (n=tony@
02:29:58 Quit obo ("bye")
02:38:12 Join YoungProGuru [0] (
02:42:52 Quit mf0102 ("Ex-Chat")
02:43:20 Join csc` [0] (n=csc@archlinux/user/csc)
02:44:09 Quit einhirn_ (Read error: 104 (Connection reset by peer))
02:46:48 Quit csc` (Remote closed the connection)
02:48:39 Join spiorf [0] (
02:54:05saratogajhMikeS: around?
02:54:32jhMikeSsaratoga: for a bit longer
02:55:40saratogacan you look at a pointer problem i have?
02:56:24saratogai changed a struct member from "frame_out[MAX_CHANNELS][BLOCK_MAX_SIZE * 2]" to "(*frame_out)[MAX_CHANNELS][BLOCK_MAX_SIZE*2]"
02:56:30saratogabasically just made a pointer to it
02:57:12saratogai then changed a memmove access to it from "&(s->frame_out[ch][s->frame_len])" to "&(*(s->frame_out)[ch][s->frame_len])"
02:57:25saratogabut the address that generates is wildly off
02:57:36 Join csc` [0] (n=csc@archlinux/user/csc)
02:58:18jhMikeS&((*s->frame_out)[ch][s->frame_len]) <= I think that's right
03:04:09jhMikeSthis compiles nicely: int myarray[10][10]; int (*myarray_p)[10][10]; myarray_p = &myarray; int* x = &(*myarray_p)[0][0];
03:05:50jhMikeSthis works too for the assignment, simply: int* x = myarray_p[0][0];
03:09:36saratogathanks, seems to work
03:14:22jhMikeSThinking about it I don't think the 2nd assignment is the same as the first but it means instead "int *x points to the [0][0] elemenent of a 10x10 array of pointers to a 10x10 array." :\
03:14:53 Quit spiorf (Remote closed the connection)
03:17:09*jhMikeS decides to go clarify this acedemic question for himself
03:23:24 Nick fxb is now known as fxb__ (
03:26:08 Quit Zarggg (Read error: 104 (Connection reset by peer))
03:26:25 Join Zarggg [0] (
03:31:22 Quit tvelocity (Remote closed the connection)
03:31:58saratogaif i put the output buffer from a codec into IRAM, DSP should be speed up correct?
03:37:01kugelI wonder if anyone can commit
03:37:05kugeljust some changes to configure
03:37:24kugelIt fixes some typos and an issue I introduced in r16935 (which doesn't seem to cause any errors though).
03:37:32*jhMikeS gets it now (*myarray)[y][z] = myarray[0][y][z]
03:37:47kugelmy patch, which was committed by a dev respectively
03:39:07***Saving seen data "./dancer.seen"
03:39:19jhMikeSsaratoga: it's normally the case since the DSP often processes in place in the output buffer
03:39:36kugelWould anyone be so kind? If not, I'll just open a tracker entry
03:41:35saratogajhMikeS: so test_codec won't see the speed up from putting the output buffer in IRAM?
03:41:39*jhMikeS learned today that "eventhough" is a word
03:42:38jhMikeSsaratoga: it should speed up the codec itself targets where it helps and speed up both in normal playback
03:42:44kugeltypo again i guess :(
03:44:30kugelOk, I updated it
03:44:30saratogabut if i see a 5% speedup in test codec, presumably the real speed up is somewhat larger because now DSP operations don't have to do so many DRAM accesses?
03:45:19jhMikeSsaratoga: exactly
03:46:19*jhMikeS wonders if test_codec should include DSP processing in combination or isolation to give more realistic results
03:46:31*kugel goes to open a tracker entry
03:47:13*jhMikeS doesn't know the script stuff well enough to judge anything about the patch unfortunately
03:50:47 Quit DerDome (Nick collision from services.)
03:50:48 Join DerDom1 [0] (
03:51:00 Nick DerDom1 is now known as DerDome (
03:51:37kugeljhMikeS: The first fix is a type in a comment, the second one in the output of configure −−help, and the third one deletes 8 lines of code which put my patch in there in r16935
03:52:49jhMikeS3rd is just an unused bit of test code?
03:53:13saratogajhMikeS: has anyone ever commented on what to do with the extra 32KB of IRAM on PP502X?
03:53:34jhMikeSsaratoga: there's another bank?
03:53:34kugeljhMikeS: Not even test code. I'm not entirly sure how it happend to be there
03:53:49saratogajhMikeS: yes 502X has 128KB
03:54:42kugeljhMikeS: The 8 lines seem to be a part of a function which is defined at the beginning of the file (I maybe copied it for refrences)
03:55:28jhMikeSsaragtoga: so the two banks aren't contiguous?
03:55:37saratogano they're contiguous
03:56:00kugeljhMikeS: But I definitlely know that these lines are wrong there and not needed
03:56:04saratogatheres just another 32k bytes at the end of the 96k on the 5020
03:56:37saratoga(and i don't believe we're using it, at least allocating it in a codec does not cause any problems once the changes are made in the lds file)
03:56:53jhMikeSsaratoga: don't use the same IRAM for codecs and plugins therefore IRAM-using plugins can run with music playing (seems the most user-visible benefit)
03:57:32saratogai thought plugins and codecs always shared IRAM?
04:03:58 Quit DerPapst (Read error: 113 (No route to host))
04:04:13jhMikeSMCF5250 iirc has 128KB and is just divided 64K for core/64K for codecs and plugins
04:05:05saratogaplaying around with that extra IRAM, I can squeeze another 6% plus whatever putting the output buffer in IRAM gets for no real effort
04:05:10saratogafor WMA
04:06:22saratogai would rather like to have all that extra IRAM for codecs unless theres some pressing need in the core for it
04:06:46saratogai think if we're ever going to beat the OF on most of the PP targets, codecs are going to need to get faster, and having more IRAM would help with that
04:11:50 Quit waldo ("Konversation terminated!")
04:14:21jhMikeSsome driver design changes (more sleep time) + voltage scaling (I seem to recall hearing that as being a possibility on PP)
04:15:08kugelWow, study mode unter MajorChanges
04:23:28saratogajhMikeS: voltage scaling doesn't really help
04:23:44saratogathe OFs don't even bother with it
04:26:59saratogai think the biggest thing is just to get CPU usage down in codecs
04:27:16saratogacommercial decoders claim things like 20MHz MP3 decoding
04:27:26saratogawe barely manage 35 or 40MHz
04:30:52jhMikeSyeah, but they use the COP
04:31:22jhMikeSsplitting DSP and decoding between cores should allow that easily
04:31:50saratogai think Buschel said the DSP only uses about 2MHz if you have all processing turned off
04:34:47kkurbjunsaratoga, are completely set on the idea that extra iram is the only solution?
04:35:20 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
04:40:01 Quit krazykit (Remote closed the connection)
04:44:06 Join krazykit [0] (n=kkit@
04:44:14 Join miepchen^schlaf_ [0] (
04:45:15saratogakkurbjun: no, but its certainly a good way to go about it, at least on the targets that have it
04:48:23jhMikeSsaratoga: but the codec is sharing CPU with other tasks so it could only really run with 100% core available to it on COP
04:48:51kkurbjunI agree, I think a clean solution to implementing flexible use needs to be come up with though, otherwise it's not a productive use of time to simply /muddy/ the code when there are other things that should and could be fully looked into as jhMikeS is suggesting. Granted these are things that come from a hardware perspective rather than code performance, but getting an answer here on what to do isn't really a consensus. I think
04:49:00kkurbjunthat or we need a steering board.
04:50:15kkurbjunthe other problem is that iram changes don't really benefit the code as a whole except on particular SOC's
04:52:15 Quit miepchen^schlaf (Connection timed out)
04:54:52soapThe code as a whole beats every other SOC OF's runtime, does it not?
04:55:32kkurbjunit doesn't beat the gigabeat OF from what I've heard
04:55:56kkurbjunI have not done a measurement on it though
04:56:13kkurbjunagain, I'm not aguing against using the IRAM, but I think it needs to be done cleanly
04:56:15soapI haven't seen any measurements on that target, solid point.
04:56:34soapBut RB doesn't scale on the F/X.
04:57:07kkurbjunsoap, generally frequeny scaling on silicon is a linear change in terms of power
04:57:14kkurbjunvoltage is exponential
04:57:27kkurbjunand those are the main factors you can change
04:57:56kkurbjunon the F we disconnect the core when code doesn't need to be run thanks to some clever code
04:58:47kkurbjunso slowing down the core only makes the code take longer to run just at a lower frequency, so it ends up pretty even
04:59:30soapI was under the impression scaling had been attempted, but didn't change a thing?
04:59:36soap(runtime wise)
04:59:41kkurbjunsoap, yep, it was the same
04:59:54 Join InigoMontoya [0] (n=621cdcfd@gateway/web/cgi-irc/
04:59:58InigoMontoyahello all
05:00:25kkurbjunit's less power while at a lower frequency, but the core has to be connected longer
05:00:37InigoMontoyai was just testing the suggested fix in FS #8641, e200 doom 90 degree crash, and it seems to have worked
05:02:03InigoMontoyahow do i go about making a patch?
05:03:48kkurbjunI guess I should clarify, voltage is not exponential, but quadratic
05:07:31InigoMontoyais the command something like diff original.c modified.c > patch.diff ?
05:08:49kkurbjun"svn diff > file.diff" will give you the differences in a file
05:09:43InigoMontoyai have modified a file, and i want to make a patch that would modify an original file to make it like my modified file
05:10:35ReKleSSInigoMontoya: you would normally use diff -u
05:10:59ReKleSSit includes some context so that the patch can still apply against a slightly different file
05:12:00InigoMontoyaReKleSS: thanks, i'll tinker around a bit and try to make a patch and post it after a little bit more testing
05:39:09***Saving seen data "./dancer.seen"
05:49:24 Join AndyI [0] (n=pasha_in@
05:50:31 Quit AndyIL (Read error: 113 (No route to host))
05:53:43 Quit YoungProGuru (Read error: 104 (Connection reset by peer))
05:56:36InigoMontoyaReKlESS: i have made the patch after testing it for awhile, what is the etiquette on posting patches? should I just post it?
05:56:54 Quit kugel ("ChatZilla 0.9.82 [Firefox 3.0b5/2008050509]")
05:57:06ReKleSSthat part I'm not sure about... I believe that's usually fine, though
06:12:44InigoMontoyaok the patch is posted in FS #8641 hopefully i did it right
06:18:40 Quit AndyI ()
06:20:31 Join YoungProGuru [0] (
06:22:01 Join AndyI [0] (n=pasha_in@
06:29:05 Quit saratoga ("CGI:IRC (Ping timeout)")
06:41:26 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
06:51:39 Nick YoungProGuru is now known as Mackes (
06:53:59 Quit Mackes ("Ex-Chat")
06:54:12 Join YoungProGuru [0] (
06:55:18 Quit InigoMontoya ("CGI:IRC (EOF)")
06:55:40 Quit YoungProGuru (Read error: 104 (Connection reset by peer))
06:55:53 Join mackes [0] (
06:56:15 Quit AndyI (Read error: 104 (Connection reset by peer))
07:09:51 Quit Anakin (Read error: 110 (Connection timed out))
07:25:16 Quit Zom (Remote closed the connection)
07:25:27 Join Zom [0] (
07:39:13***Saving seen data "./dancer.seen"
07:42:40 Join dabujo [0] (
07:47:46 Join RoC_MasterMind [0] (
07:48:22JdGordonis the microdrive in the mr100 replaceable with a CF card?
07:51:52homieloweI think its the same seagate type as the ipod mini
07:53:31 Join Buschel [0] (
07:54:31*Buschel reads the logs and support saratoga in his statement about CPU-usage of OFs
07:54:59Buschelseems like mp3-decoding on arm7 (non dual core) is possible @20-24MHz.
07:55:32Buschelthis is equal to mpc in rockbox −− with mpc-playback we surpass the OF battery runtime...
07:55:45Buschelso: let's tweak the decoders! ;o)
08:05:48 Quit Buschel ()
08:07:01 Join Mathiasdm [0] (
08:38:21 Join crash91 [0] (n=crash@
08:38:58 Join Rob2223 [0] (
08:41:19 Nick miepchen^schlaf_ is now known as miepchen^schlaf (
08:51:55 Quit Crayzie (Read error: 110 (Connection timed out))
08:56:41 Quit Rob2222 (Read error: 110 (Connection timed out))
08:59:47 Quit crash91 ("Ex-Chat")
09:02:00 Quit RoC_MasterMind ("Leaving")
09:02:15JdGordonanyone know if low_light is ever on irc?
09:09:11scorchelow_light (i=c730180a@gateway/web/ajax/ signed off 1 day and 12 hours ago
09:09:34 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
09:10:18*JdGordon does some ESP thing to get him to come back online some time tonight :p
09:11:57scorcheESP is mainly RX...not TX
09:12:25JdGordondamn you!
09:12:41JdGordonand how can it only be RX? something has to TX...
09:19:30zewhat you want is telepathic suggestion
09:20:16KohlrabiIs the latest version of rockbox able to decode Musepack SV8?
09:20:19zeESP can almost be thought of as a special kind of filter... it filters out immediate sensory input along with the effect of obstructions in order to recieve sensory perception from outside your immediate scope
09:21:12KohlrabiFlkyspray says "no", but maybe it has changed :I
09:22:58scorcheze: wrong channel ;)
09:23:24zescorche: i know, i was done :p
09:23:56 Part toffe82
09:28:32 Join n1s [0] (n=nils@rockbox/developer/n1s)
09:32:30*n1s sees that the string order in english.lang was broken without bumping the version or even a warning...
09:39:16***Saving seen data "./dancer.seen"
09:41:49 Quit csc` ("If you can't do anything smart, do somethin right ~Serenity")
09:54:49 Join JdGordon|uni [0] (i=d31c5d08@gateway/web/ajax/
09:56:38 Join webguest12 [0] (n=d9e1fbe6@gateway/web/cgi-irc/
09:56:56 Join Buschel [0] (
09:57:34webguest12hi, if I have rockbox on my Archos JBR will a apple mac see it ?
09:58:02n1swebguest12: yes
09:58:25webguest12I dont need special drivers for the mac ?
09:58:50scorchewebguest12: which JBR?
09:59:10scorcheand what size?
09:59:28webguest12I modded mine to 12o gigs
09:59:44scorchei meant what size originally
10:00:01webguest12hmm, 10 or 15 gig originally
10:00:25scorchewell...let me ask it USB 1.1 or 2.0?
10:01:46scorchei dont know the situation with those drivers and a mac...does it work fine currently?
10:02:06JdGordon|unishouldnt it just show up as a regular mass storage device?
10:02:24scorcheJdGordon|uni: 1.1 required drivers
10:03:02webguest12so I'll need drivers for the mac ro see it ?
10:03:07 Quit JdGordon|uni (" ajax IRC Client")
10:03:16scorchewebguest12: does the mac currently see the device?
10:04:05webguest12haven't tried it yet, mac is a friends pc,
10:04:45scorchewell, rockbox wont make a difference in regards to if the mac will see it
10:05:01webguest12ah, that answers it then, thanks
10:14:13 Join desowin [0] (n=desowin@atheme/member/desowin)
10:14:22 Join davina [0] (
10:22:55 Quit miepchen^schlaf ()
10:24:10webguest12scorhe; do you perhaps have the original installation CD's for the Archos JBR V1 ?
10:25:04scorcheprobably, but i would probably have to spend hours of digging through old boxes
10:25:52webguest12ok, I bought my Archos off ebay, so no instalation disk, apparenlt it had a mac installation disk
10:26:06*scorche nods
10:26:30webguest12wanna sell me the disk :-)
10:26:39scorchei dont have it handy
10:27:31webguest12i can wait a few days if it makes any difference, but if you dont to want to thats ok
10:27:43scorchei dont :)
10:27:50webguest12np :-)
10:28:00Mouser_XTry searching for the drivers.
10:28:11Mouser_XThere's, or something.
10:28:17webguest12I have, nothing for the mac only pc
10:28:30Mouser_XI meant searching online. Like Google.
10:28:36Mouser_XSorry, misread.
10:28:39Buschelkohlrabi: musepack sv8 ist still not supported yet. i've made a first attempt to compile it into rockbox yesterday −− and failed :/ will need some more time until i get it to work
10:29:16 Quit feisar ("leaving")
10:29:35Mouser_XOdd as this may sound, have you tried downloading the various drivers you've found? Maybe they contain the MAC drivers in there after all.
10:29:36 Join feisar [0] (
10:29:52Mouser_XAlso, try searching for *.hfs files, and not *.exe or *.zip.
10:30:13Mouser_X(Or whatever the MAC's "native" archive format is.)
10:31:00webguest12I'll try all that, but the packagges all state various flavours of windows, I just need to find someone with the installation disks that is willing to copy the drivers off them, thanks for your time
10:37:14 Quit desowin ("KVIrc 4.0.0 Insomnia")
10:37:25Mouser_Xwebguest12: Is this it? -
10:38:08scorcheMouser_X: keep in mind this is all before OSX
10:39:05Mouser_XI searched for MAC, not OSX.
10:39:33*Mouser_X doesn't use MACs, and thus is almost entirely unfamiliar with their stuff, including their OS.
10:39:49scorchei doubt you will find OSX drivers for it....and we dont even know if they are needed yet (i dont know what sort of drivers OSX includes)
10:39:51webguest12thank you for your effort, but it is indeed for the OSX
10:40:11Mouser_XAh, sorry.
10:40:16*Mouser_X goes to bed.
10:40:58scorchewebguest12: see if it works...if not, come can always wait around till someone else comes around that might know too
10:41:19webguest12thank you, I'll do so
10:51:38linuxstbwebguest12: This may be what you're looking for. IIRC, it's the ISD200 USB bridge chip you need a driver for -
10:54:01 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
10:59:35 Join GodEater [50] (n=bryan@rockbox/staff/GodEater)
10:59:52 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
10:59:59 Join obo [0] (n=obo@rockbox/developer/obo)
11:08:48 Join desowin [0] (n=desowin@atheme/member/desowin)
11:08:56 Join ender` [0] (
11:09:47 Join gregzx [0] (
11:13:57 Quit mcuelenaere ()
11:16:45 Join bertrik [0] (
11:20:34 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
11:21:40KohlrabiBuschel: good to know, thanks :)
11:25:14 Join waldo [0] (
11:37:34JdGordonNico_P: done anything with the icon list view yet?
11:37:48Nico_PJdGordon: not yet, no
11:38:01 Quit BHSPitLappy (Remote closed the connection)
11:38:01Nico_PI've done some basic thinking but no coding
11:39:20***Saving seen data "./dancer.seen"
11:39:35 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
11:41:37Nico_PI hope to start coding soonish
11:44:12 Join PaulJam [0] (
11:47:23 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
11:50:15 Quit ReKleSS (Remote closed the connection)
11:52:35 Nick fxb__ is now known as fxb (
11:54:08 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
11:54:16 Quit linuxstb (Nick collision from services.)
11:54:22 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
11:54:38 Join tvelocity [0] (
11:56:56JdGordonNico_P: got a min to checkout a wps/playback bug?
11:57:55JdGordontrying to see if 6078 is still a bug.. The A-B repeating works correctly (i tihnk), but the statusbar shows wrongly
11:59:01JdGordonit colours the first bit of the bar instead of starting from the a-b spot
11:59:01 Quit homielowe ()
11:59:38JdGordonalso, the possibly bigger issue is that it finishes the track after you set the B spot then jumps back to A as expected, but should it jump back as soon as the B is set?
11:59:47*JdGordon will try looking at the progressbar issue
12:03:57 Join PaulJam_ [0] (
12:10:15JdGordona-b mode doesnt seem to work in the sim at all?
12:11:10Nico_PJdGordon: that's a limitation of A-B repeat
12:11:20JdGordonwhich part?
12:11:43Nico_Prepeating the wrong section (looking at FS #6078)
12:11:49Nico_Pit's because of PCM latency
12:12:06Nico_Pabout the progressbar, I'm not sure I understand the issue
12:15:35JdGordonNico_P: Its not repeating the wrong section (well on my sansa and a 30s track anyway), but the progress bar is filling 0:00 -> 0:15 instead of say 0:10 -> 0:25
12:15:51JdGordonbut, apparently works fine on longer tracks, so i guess its a non issue
12:16:33Nico_PI think that's something that can't really be fixed without using a whole new approach to the feature
12:16:56Buschelpreglow: you there?
12:19:20 Join PaulJam__ [0] (
12:20:24 Join spiorf [0] (
12:22:28 Quit PaulJam (Read error: 110 (Connection timed out))
12:26:00 Nick PaulJam__ is now known as PaulJam (
12:29:10 Join DerPapst [0] (
12:31:29 Join petur [50] (n=petur@rockbox/developer/petur)
12:32:24 Join das-perceptron [0] (
12:32:49 Part das-perceptron
12:36:58 Quit PaulJam_ (Read error: 113 (No route to host))
12:41:26 Join cnakihlcm [0] (i=0@
12:45:12 Join |desowin| [0] (
12:46:14 Quit |desowin| (Client Quit)
12:47:08 Join barrywardell [0] (n=barrywar@
12:59:28 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
13:00:52 Quit desowin (Read error: 110 (Connection timed out))
13:03:17 Join desowin [0] (n=desowin@atheme/member/desowin)
13:05:48 Join miepchen^schlaf [0] (
13:07:54 Quit bertrik ("bye")
13:08:44 Join bertrik [0] (
13:09:55 Quit Xerion (" ")
13:11:34 Quit merbanan (Remote closed the connection)
13:11:53 Join ali_as [0] (
13:14:02JdGordonn1s: nice commit :)
13:19:01bertrikdo we really need to be that strict w.r.t. line length (80 characters I presume)?
13:19:29JdGordondepends who you ask :p
13:19:32JdGordonand yes, its 80chars
13:19:46scorcheit ought to be enough for anyone!
13:25:28webguest12linuxstb: thanks for the link but it has a weird redirect to it and I'm unable to download it, I'll try and source it from elsewhere
13:27:30webguest12damn all th links for Cypress OS X Storage Driver - 1.2 point to the same page and it gets redirected quicly, very annoying
13:28:41 Quit GodEater (Read error: 113 (No route to host))
13:29:00 Join GodEater [50] (n=bryan@rockbox/staff/GodEater)
13:29:08linuxstbwebguest12: I assume you've tried attaching your Archos to your Mac?
13:29:52*linuxstb is failing to create a modified firmware that the Clip will accept...
13:30:01 Quit ali_as_ (Read error: 110 (Connection timed out))
13:30:37webguest12no I dont have a mac, a friend who has a mac wants my Archos player but obviously he wants drivers for USB V1.1
13:31:26domonokylinux has those archos usb drivers included, but i dont know about mac ?
13:31:39webguest12I have no idea what OSX needs with regards to Archos JBR v1 usb1.1 drivers
13:31:41linuxstbI would simply try it - as domonoky said, maybe the drivers are included...
13:33:02webguest12hmm, yep but i want to be prepared for what if it doesn't, windows needs specific drivers so I'm assuming that OSX does as well, or is apple that good ? :->
13:37:26 Join Thundercloud [0] (
13:39:22***Saving seen data "./dancer.seen"
13:40:44 Quit gevaerts ("the rest of the day is reserved for RL")
13:43:28 Quit mcuelenaere (Read error: 110 (Connection timed out))
13:48:02 Join nplus [0] (
13:58:12 Join Xqtftqx [0] (i=40e9e3d2@gateway/web/ajax/
13:58:41XqtftqxHey, does anybody know of a program that will export rss feeds to text? i want to run it off my gigabeat to get rss feeds
14:01:12 Quit petur ("also have the rest of the day for RL ;)")
14:03:06XqtftqxOk... I will just make my own then.
14:03:09 Join fdinel [0] (
14:07:48 Join krazykit_ [0] (n=kkit@
14:08:38 Quit Xqtftqx (" ajax IRC Client")
14:10:20 Quit dabujo (Read error: 104 (Connection reset by peer))
14:11:15 Join hannesd [0] (
14:13:18 Quit BigBambi (Remote closed the connection)
14:15:53 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
14:16:42 Join PaulJam_ [0] (
14:18:10 Join barrywardell_ [0] (n=barrywar@
14:20:15 Quit krazykit (Read error: 110 (Connection timed out))
14:22:30 Join GodEater_ [50] (n=bryan@rockbox/staff/GodEater)
14:22:32 Quit GodEater (Read error: 113 (No route to host))
14:23:22 Join PaulJam__ [0] (
14:24:02 Quit PaulJam (Nick collision from services.)
14:24:08 Quit PaulJam_ (Nick collision from services.)
14:24:12 Nick PaulJam__ is now known as PaulJam (
14:27:48 Quit barrywardell (Read error: 110 (Connection timed out))
14:28:48 Join Lear [0] (i=chatzill@rockbox/developer/lear)
14:30:26 Quit nplus ("Leaving")
14:30:31 Join nplus [0] (
14:46:00 Join Xqtftqx [0] (i=40e9e3d2@gateway/web/ajax/
14:48:32 Quit tvelocity ("Αποχώρησε")
14:54:34 Quit gregzx ("ChatZilla 0.9.82 [Firefox]")
14:56:50 Join Evilnick [0] (
15:02:41 Quit Xqtftqx (" ajax IRC Client")
15:13:06 Quit webguest12 ("CGI:IRC")
15:21:09 Join gregzx [0] (
15:27:23 Quit Lear ("ChatZilla 0.9.82 [Firefox 3.0pre/2008050906]")
15:27:26 Quit PaulJam (Read error: 110 (Connection timed out))
15:28:32 Quit DerPapst ("*plop*")
15:34:00 Quit fdinel (Read error: 104 (Connection reset by peer))
15:37:20 Join Mathiasdm2 [0] (
15:37:23 Join Arathis [0] (
15:37:34 Quit Mathiasdm (Read error: 104 (Connection reset by peer))
15:39:25***Saving seen data "./dancer.seen"
15:46:49JdGordon*phew* kept the table green :D
15:47:02JdGordonapart from the not green bits :p
15:48:46JdGordonlinuxstb: can set_viewport(), clear_viewport() and stop_Scroll() all be merged into 1 call?
15:49:39JdGordonand would there be any point doing that? or would it use just as much bin doing that as it would just doing the 3 calls seperatly
15:54:03 Join mf0102 [0] (n=michi@
16:00:08JdGordonjust radio and the alarm screen still using textarea stuf :)
16:02:20JdGordonNOOOOOOOOOOOOOO!!!!!!!!!!!!!!! forgot to fix the #include!
16:03:03 Join SirFunk [0] (
16:04:00linuxstbJdGordon: Doesn't clear_viewport call stop_scroll already?
16:04:23JdGordonI dont know...
16:04:26 Quit SirFunk (SendQ exceeded)
16:04:58 Join Xqtftqx [0] (i=40e9e3d2@gateway/web/ajax/
16:05:01linuxstbYes, it does. (checking the code)
16:05:04 Join SirFunk [0] (
16:05:05*JdGordon too slow
16:09:58 Quit Ragnor (Nick collision from services.)
16:10:09 Join Ragnor [0] (
16:11:45 Join Clix2 [0] (n=568dbdd3@gateway/web/cgi-irc/
16:12:22Clix2Hey. does anyone know if there is a rockbox firmware upgrade so that the Clix 2 can read FLAC files?
16:12:32JdGordonthere isnt
16:12:43Clix2oh ok
16:12:56Clix2Are there any plans for doing it?
16:13:14JdGordonports arnt planned... they just happen (if they happen)
16:13:49Clix2oh right, ok. FInally, do you know of any other.. program of sorts that allows that?
16:14:30JdGordonno, but isnt the clix 2gb? wouldnt mp3/ogg be better for it anyway?
16:14:39linuxstbYes, a program that converts FLAC to a format the Clix can play...
16:14:46Clix2no, it's the clix 2 8gb
16:15:12Clix2linux, but the point of me wanting FLAC is just because it's lossless..
16:15:30JdGordonthen no.. your out of luck
16:15:53Clix2ok. thx for your help
16:17:59 Quit Clix2 ("CGI:IRC (EOF)")
16:28:57XqtftqxGuys, Can i run a gigabeat F theme on the S?
16:29:09Xqtftqxthe screen size is the same, i think
16:29:38XqtftqxIll try it in the simulator
16:30:40linuxstbYes, they should work.
16:33:14Xqtftqxthey work
16:33:47XqtftqxMan, the graphics engine for the gigabeat needs some work
16:34:02XqtftqxHmmm... Im gonna look at it
16:34:07Xqtftqxsee if i can fix anything
16:35:02XqtftqxLCD Driver*, Sorry
16:35:20linuxstbWhat's the problem?
16:35:26XqtftqxIt needs work
16:35:46XqtftqxDoom wont run because its to low, RockBoy barely runs anythings.
16:36:13XqtftqxFor the S
16:39:51XqtftqxWhat theme do you like for the F?
16:47:16linuxstbI just use the default - I don't care much about themes.
16:48:06XqtftqxI am probly gonna use Cabbie, i just want to see how pen and paper looks
16:48:53XqtftqxHey i got a question, when a new bootloaders realsed, can i boot rockbox into usb mode and put the nk.bin onto the firmware partion without erasing any data?
16:50:14Xqtftqxoh linux, the backdrops work and fonts for for the F on the S but the playing screens messed up
16:50:25bertrikany lithium charging experts online?
16:51:01Xqtftqxi know some stuff
16:51:06Xqtftqxim not a expert
16:51:18 Join kugel [0] (n=kugel@unaffiliated/kugel)
16:52:00n1sXqtftqx: re bootloader: yes you can
16:52:15XqtftqxK thanks
16:52:31XqtftqxI just got to remember to use a multiboot one
16:53:16XqtftqxWow im suprised with the battery life for just playing audo with the screen off
16:53:36Xqtftqxits been about 12 hours
16:53:43Xqtftqxand its on 72
16:54:02bertrikXqtftqx: do you know if it's harmful for a lithium battery to leave it connected to a charger in voltage-limit mode, at say 4.1V?
16:54:37XqtftqxHow much was the battery ment to take in
16:54:54XqtftqxThat Number and anything below that is safe
16:55:47bertrikDepends on what? It's a 750 mAh battery (for the e200) or a 530 mAh battery (for the c200).
16:56:51XqtftqxIm not sure
16:56:57bertrikThe problem is that the charger has an end-of-charge signal, but it is only signaled when charging from partially discharged state. If the battery is already at say 4.0V, it never kicks in so the charger stays on at 4.1V (don't know the current at that point though)
16:58:23XqtftqxThats kinda odd, but do you need it equactly at 100% charged?
16:59:31bertrikI guess 90% is OK too, especially if this lessens the wear on the battery
16:59:49domonokybertrik: isnt charging handled in hardware on a c200 or e200 ?
17:01:40bertrikdomonoky: partially it seems, there is a charging circuit with current and voltage limits, under/overvoltage protection, high-temperature protection. But the datasheet also says to turn off the charger whenever a certain end-of-charge bit becomes active. The problem is that this bit doesn't always become active even after charging for very long.
17:01:50XqtftqxHey guys, on the main page, i see things like "fixed yellow" what does this mean?
17:02:22bertrikXqtftqx: it means there were compiler warnings which result in a yellow cell in the build table
17:02:24domonokyXqtftqx: yellow means a warning of the compiler..
17:03:27XqtftqxWheres the build table?
17:05:23bertrikdomonoky: In some cases, the bit does become active, but too soon it seems. I wonder now if I have to bother with that bit at all, or just keep the charger switched on (I think that the charging current eventually tapers off to 0 anyway in voltage-limit mode)
17:06:28XqtftqxGuys, can i run voice files on the S?
17:06:41domonokyi think leaving it on should be save,the charging circuit should handle it, but i am no expert..
17:06:57n1sXqtftqx: you should be able to but will have to create one yourself
17:07:15domonokyvoice should work on the S...
17:07:31n1si don't know if anyone has tested it though
17:07:34domonokybut the S is very new, so expect some errors.. :-)
17:07:43XqtftqxI know
17:07:50XqtftqxIve found many many errors
17:08:50XqtftqxWheres the wiki page for creating voice files?
17:08:54n1sactually it already works nicely for me so a bit moore battery life and i'm happy :)
17:09:13Xqtftqxdo i have to build the voice files from source?
17:09:25n1sXqtftqx: what os are you on?
17:09:37XqtftqxBut ive got a vm if i need it
17:09:52domonokyhow is bootloader installation for Gigabeat S going ? is there a method for windows ? (thinking about rbutil support)..
17:10:08n1sXqtftqx: i'm no help then...
17:10:16Xqtftqxi put the bootloader on through windows
17:10:21domonokyXqtftqx: you can create voicefiles with the buildsystem in cygwin on windows.. (and use spai5 voices)
17:10:30XqtftqxOk ill try
17:10:47XqtftqxI chouldnt Download like the gigabeat F voices?
17:10:57XqtftqxWhats the Diffrence?
17:11:26domonokythe language files are different (different strings) so also differen voice file..
17:11:28 Join Crayzie [0] (
17:11:33n1sthere are different clips for different players so voice files need to be different
17:11:53n1salso there's a mechanism to prevent you from loading the wrong one
17:12:08domonokyso if you use the voicefile for a f on a S,you will hear sound, but it will speak wrong menu entrys..
17:12:18n1sdomonoky: no it will not load
17:12:22XqtftqxIt wont
17:12:28XqtftqxI tried it earlier today
17:12:40XqtftqxIs there a wiki page for creating voice files?
17:13:04n1sXqtftqx: you could try searching
17:13:13XqtftqxI tried and chouldnt find anything
17:13:17domonokyXqtftqx: its really easy with the buildsystem, just configure for voice, and run "make voice" ..
17:13:29XqtftqxOh realy?
17:13:37Xqtftqxill go and take a look
17:13:38*n1s uses google ->
17:13:57XqtftqxLol thanks
17:14:26domonokythis one is better.. :-)
17:14:43n1syeah but that's linked from the one i posted ;)
17:14:48XqtftqxThat may help a little more
17:15:30XqtftqxDang, i have to make my own wav files
17:16:01n1sXqtftqx: that's what you use a tts engine for
17:16:03domonokyXqtftqx: read better... there are TTS engines..
17:16:16XqtftqxI was getting there...
17:17:50 Quit hannesd (Read error: 113 (No route to host))
17:20:57XqtftqxAwwww... Its not worth it,
17:21:29*n1s got curious and tries it ;)
17:21:37n1son the beast that is
17:21:49Xqtftqxcan i have a copy if you dont mind?
17:21:57XqtftqxPlus im running vista
17:23:12n1shmm, warnings in wavtrim and voicefont
17:23:33Xqtftqxtheres warnings in a whole bunch of stuff for the S
17:24:16n1sXqtftqx: these are general utilities and the same for every target but where are there warnings for the S ?
17:24:43XqtftqxI get them all the time when compiling
17:24:56Xqtftqxits probly because i dont have the apps
17:25:02Xqtftqxi use the rockbox vm
17:25:11Xqtftqxi can compile rockbox
17:25:13Xqtftqxbut not the bootloader
17:26:50n1syup, voice works fine on the beast
17:28:10 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
17:29:53*jhMikeS just enabled write buffer on the device memory section in the S and it helps immensely with frame speed (no dropped mpegplayer frames)
17:31:02Xqtftqxthanks n1s
17:31:14jhMikeSXqtftqx: Thanks for the heads up. crt0.S memory setup wasn't complete.
17:31:26XqtftqxWhat heads up?
17:31:54XqtftqxFor the LCD driver?
17:32:36jhMikeSthe driver code is fine - it was the memory setup.
17:32:59jhMikeSjust had to enable write buffering to the 1MB section
17:33:12Xqtftqxis it in the svn?
17:33:43n1sjhMikeS: that's great :) do we use the l2 cache in the imx31?
17:34:12XqtftqxAlso the sound in rockboy is horrid!
17:34:27n1sXqtftqx: that is not specific to the S
17:34:44XqtftqxIt was fine on my F
17:35:09 Quit Horscht (Read error: 110 (Connection timed out))
17:35:14jhMikeSn1s: yes
17:35:18 Join Horscht [0] (
17:35:20n1sit's horrible on h300 at least... but maybe it has to do with speed
17:35:20jhMikeSXqtftqx: shortly
17:35:30jhMikeSand it will need a bl replacement
17:35:42 Join tootoo_ [0] (
17:35:46XqtftqxjhMikeS: Is it in the memory setup thing in the svn?
17:36:08jhMikeSXqtftqx: no, I'm going to commit the change
17:36:28XqtftqxOk, will that fix the doom plugin?
17:36:32 Part tootoo_
17:38:30XqtftqxHey jsMikeS can you compile me a build with the fix? ive been having some troubles building
17:38:56BigBambiXqtftqx: you will see when things get fixed by the commits on the front page. Please stop bothering people with pointless questions, and for the last time PLEASE DO *NOT* use the enter key as punctuation
17:39:29***Saving seen data "./dancer.seen"
17:39:55BigBambiXqtftqx: If you read the IRC logs for the last few days, there is somewhere you can download the SVN S build from, it just isn't linked to from the site yet, as the port isn't ready for general use
17:39:56XqtftqxSorry, ill stop using the enter key, i promise
17:41:34XqtftqxIsnt it on the install page?
17:41:37 Quit Mathiasdm2 ("Yuuw!")
17:42:12BigBambiAlso, the fact that things like rockboy and doom do not work properly yet is of low importance - Rockbox is meant primarily for music playback
17:43:01BigBambiXqtftqx: That is what I meant, but it isn't linked to from anywhere, such as
17:45:44 Quit Xqtftqx (" ajax IRC Client")
17:49:35 Join Xqtftqx [0] (i=40e9e3d2@gateway/web/ajax/
17:49:35 Join Lucidiot [0] (n=john@unaffiliated/lucidiot)
17:50:26XqtftqxBibBambi its on the gigabeatSInstall wiki page
17:50:27 Join _sourcemaker [0] (
17:50:33 Part _sourcemaker ("Kopete 0.12.5 :")
17:52:40Xqtftqxn1s, to enable the voice i put the file in the lang dir and go into settings>General>voice>Voice Menus=yes correct?
17:52:54XqtftqxOh wait, its working now
17:52:57 Part Lucidiot
17:55:04 Join hannesd [0] (
17:55:53XqtftqxGuys, how did compile the multiboot bootloader? I have the orginal nk.bin
17:56:04BigBambiyou use mknkboot
17:56:29Xqtftqxmknkboot nk.bin(orginal) nk.bin(rockbox)
17:56:48n1sXqtftqx: how to patch nk.bin with the rockbox bootlader is described in the GigabeatSInfo page in the wiki
17:56:50BigBambisee the wiki page
17:57:09XqtftqxOk thank you
17:57:47XqtftqxSorry but, where is it on that page?
17:58:04BigBambiXqtftqx: why not just read it?
17:58:42 Quit mf0102 (Remote closed the connection)
17:58:47BigBambiIt is a linux tool, so I suspect under the installation from linux section
17:58:57 Join mf0102 [0] (n=michi@
17:59:00BigBambiOr rather, loading code from linux
18:00:39Xqtftqxim looking at that
18:00:46 Quit ompaul (Client Quit)
18:00:54 Quit mf0102 (Remote closed the connection)
18:01:52Xqtftqxoh wait i see, i was confusing rockbox bootloader and eboot.bin
18:02:09BigBambieboot.bin is nothing to do with it
18:04:27Xqtftqxso the rockbox nk.bin is considered the boot file?
18:05:18BigBambink.bin produced by building the bootloader is the rockbox bootloader
18:05:29XqtftqxOk, thanks
18:05:57soapWorth my while to update the iPod bootloader on my devices?
18:06:11soapI haven't updated since the "must" upgrade.
18:06:14BigBambiXqtftqx: If you want to use mknkboot to patch the original nk.bin to produce a dualboot bootloader, you need to use bootloader/bootloader.bin, as that page says
18:07:08BigBambiXqtftqx: the nk.bin that is built is the rockbox bootloader only, it is not dual boot
18:07:17Xqtftqxi know
18:08:40n1sany flyspray admin people around? recently i've been getting a blank page when trying to search for "All task types"
18:09:40XqtftqxOk so im just checking, the link above ( already had the 1MB patch thing?
18:09:46BigBambisoap: which must upgrade?
18:10:03BigBambithe newest one yes, it depends when you downloaded it
18:10:47BigBambi@ Xqtftqx
18:11:03XqtftqxOk, im gonna download it now
18:11:16 Quit desowin (Read error: 113 (No route to host))
18:11:26soapBigBambi, the "must" upgrade over a year ago when you had to upgrade.
18:11:41BigBambiXqtftqx: you don't need to tell us everything you are doing. This is a logged development and support channel, and we would like to keep the logs as clean as possible
18:11:51BigBambisoap: In that case, I'd say yes :)
18:12:02soapwhat do I gain?
18:12:13BigBambiI think you just run ipodpatcher again
18:12:17BigBambior rbutil
18:12:46BigBambiheh, disregard that, I read "What do I do again" and thought it was a little joke :)
18:13:52BigBambisoap: I'm sorry, I seem to have misunderstood - I read "I haven't updated since the "must" upgrade." as not having upgraded since before it
18:14:02BigBambiWhich thinking about it can't be true for you
18:14:02 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
18:14:18BigBambiAnd so if it is already that, I don't know if there is any point
18:16:24*jhMikeS screams at powermgmt.c
18:16:45XqtftqxI know, stupid thing wont charge inside rockbox!
18:16:58jhMikeSI know it won't
18:17:14BigBambiXqtftqx: jhMikeS is the main dev working on the beast
18:17:15jhMikeSit's an #ifdef hell
18:17:15linuxstbsoap: I can't recall any useful bootloader changes for the ipod - unless you want the bootloader logo...
18:17:36BigBambiXqtftqx: but powermgmt.c is for other targets too
18:17:48XqtftqxOh, sorry
18:17:51jhMikeSthe beast needs a completely new algorithm and it's not accomodating to adding that
18:17:53linuxstbjhMikeS: You're not the first to observe that... I seem to recall amiconn saying he wanted to rewrite it
18:18:09XqtftqxI have to switch into toshiba firmware to charge
18:18:30BigBambiXqtftqx: We know. PLEASE stop pointing out the known on a not-finished port
18:18:36BigBambiIt is getting very tiresome now
18:18:47XqtftqxSorry, i wasnt telling anybody
18:18:49jhMikeSI might have to do a fresh write and see about merging back upwards
18:18:58 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
18:19:13BigBambiXqtftqx: this isn't a social channel, unless you have something useful to add, don't
18:19:35XqtftqxAlright, ill stop
18:20:42XqtftqxjhMikeS: Acullary The S charges a little inside rockbox.
18:20:52BigBambiXqtftqx: !!!!!
18:20:56BigBambiHE KNOWS
18:21:24jhMikeSif anyone wishes to figure out how/where the framebuffer address is setup it would be great because the memory mapping isn't flat because of it being at 1MB in physical RAM. Gotta be in the IPU-SDI somewhere.
18:21:33 Join toffe82 [0] (
18:22:03soaphmm, bootloader logo, let me look at that bitmap.
18:22:10*bluebrother notices conflicts in his rbutil folder :/
18:24:45barrywardell_soap, linuxstb there's no bootloader logo in the ipods (yet?)
18:25:06linuxstbAh no...
18:25:23bluebrotherdomonoky: regarding rbzip.cpp: wouldn't it be better to make it like rbunzip −− don't depend on the logger but only emit signals which are connected outside?
18:25:31barrywardell_it would be easy to add if that's desired...
18:25:47 Join fml [0] (n=4fd3f3fb@gateway/web/cgi-irc/
18:26:33linuxstbbarrywardell_: I wouldn't go as far as saying I desire it, but we may as well add it.
18:26:36soapwith glass effects and reflections!
18:26:53fmlWhen my Sansa c250 boots I can see the line 'Model: e200'. Is that correct? Isn't that confusing?
18:29:02 Quit miepchen^schlaf ()
18:29:12 Quit Arathis ("Bye, bye")
18:30:21barrywardell_fml: do you mean 'Model id: e200'?
18:31:38 Quit Zarggg (Read error: 104 (Connection reset by peer))
18:32:12fmlbarrywardell_: probably yes, I don't have the device at hand now.
18:32:57Xqtftqx"" can somebody look at those errors
18:33:18XqtftqxSorry, guys im trying to put a link, to a image,
18:33:30Xqtftqxput a ".png" on the end of that
18:33:39BigBambiXavierGr: please paste the text output into something like
18:33:48BigBambisorry, Xqtftqx
18:33:57 Join Zarggg [0] (
18:34:03barrywardell_fml: strange. I can't see why that would be from looking at the code
18:34:54linuxstbCan anyone explain what I'm doing wrong with this ARM assembly?
18:35:00fmlbarrywardell_: but do you also find it confusing? Or is something different meant by this string?
18:35:15XqtftqxI need some help with those errors
18:35:30BigBambiXqtftqx: please paste the text output into
18:35:52Xqtftqxhow can i copy it from the vm?
18:36:05fmlbarrywardell_: and another question: when booting Sansa e200, I see no text output, but for c200 I see some. Is it possible to hide it and to show the logo only?
18:36:15BigBambiXqtftqx: you need vmtools, check google
18:36:58barrywardell_fml: i think it should be c200, but can't see why not from looking at the code
18:37:26fmlbarrywardell_: what file is it?
18:37:48barrywardell_fml: what text output do you see? the booting should be fairly identical on e200 and c200
18:38:28barrywardell_fml: line 1535 and 1542 of tools/configure
18:39:44 Join csc` [0] (n=csc@archlinux/user/csc)
18:39:54fmlbarrywardell_: does the bootloader for e200 also work for c200?
18:40:19linuxstbIt's a different LCD driver for a start.
18:40:49fmlI 'infected' my c200 with rockbox via RbUtil. Could it be that the wrong bootloader was installed?
18:40:57fmllinuxstb: ah...
18:41:23toffe82for the ones interested, I upload the block diagram of the philips gogear 6330 on the wiki
18:43:04fmlbarrywardell_: I can't tell exactly what output it is since I dont' have the player now. But I remember very clearly that I've seen e200 there.
18:47:11barrywardell_fml: i think you should try updating to the latest bootloader and see if that fixes things
18:47:24 Quit daskhweg (Read error: 104 (Connection reset by peer))
18:47:32XqtftqxFailed to merge target specific data with libgcc.a
18:47:42XqtftqxWhat does that mean? i keep getting that error
18:47:50barrywardell_fml: also, there hasn't actually been a bootloader released with the boot logo
18:49:04Buschelhmm, there are lots of flyspray entries doubled. like all the feature requests for pitch...
18:49:21Buschelshall close of the doublettes?
18:49:38Buscheloops, an "I" was missing
18:49:39bluebrotherBuschel: go closing them ;-)
18:49:56bluebrotherduplicates are boring and annoying ...
18:50:19linuxstbXqtftqx: It means you're running the wrong arm-elf-gcc - you need one with the "multilibs patch". The current script will build one.
18:50:33Buschelalso: FS #1889 (mpeglpayer scrollbar) is implemented, correct?
18:50:59Xqtftqxwheres the in the svn?
18:51:14BigBambi /tools/
18:51:19Xqtftqxk thanks
18:51:27linuxstbfind . -name
18:52:36Xqtftqxwhat do i do with (sorry im confused)
18:52:38linuxstbBuschel: That's not mpegplayer, it's the video plugin for Archos devices
18:53:02*Buschel was too fast then
18:53:15linuxstbBuschel: You only asked the question 2 minutes ago...
18:55:07jhMikeSFS #1889? I don't think mpegplayer existed before FS #5k-6k. ;)
18:55:46linuxstbYes, task opened in 2003...
18:55:58n1sXqtftqx: you run it...
18:56:00*Buschel just again was reminded to read precisely
18:56:07Xqtftqxit says command not found
18:56:24n1s./ before
18:56:52linuxstbjhMikeS: Can you help me with this? I'm not sure if the "bne" instruction is being compiled correctly...
18:57:04domonokybluebrother: how is your bootloader install rework going ? :-)
18:58:25bluebrotherdomonoky: well, it's progressing. Unfortunately it's rather slow.
18:58:29jhMikeSlinuxstb: ok
18:58:33Nico_Pbluebrother: re r17460 (convert to unix line endings): is this cosmetic, or is there a requirement for unix EOLs even on windows?
18:58:50barrywardell_linuxstb, soap: don't have an ipod to test on, but this shoot add the bootloader logo for ipods:
18:59:01fmlbarrywardell_: I completely installed rockbox two days ago, so I think that the BL is quite up to date
18:59:14jhMikeSNico_P: were you by any chance inclined to try without the ata lock hack? :) I wonder if amiconn's commit takes care of the problem.
18:59:16XqtftqxOk guys its downloading
18:59:47bluebrotherNico_P: this is only cosmetic. But svn should convert them to windows line endings if you check the file out on a windows box (read: using a windows build of svn)
18:59:56Nico_PjhMikeS: I could test, yeah. could you remind me where and how to disable the hack?
19:00:28Nico_Pbluebrother: you need to set svn:eolstyle for that. I'm in the proces of doing it
19:00:45barrywardell_fml: in that case, I'm not sure what's happening. i do know that the bootloaders should work the same on c200 and e200 though
19:00:55BigBambiXqtftqx: I'm really serious now - I have asked you many times, and you keep ignoring it. Don't tell us exactly what you are doing all the time. Go and read the posting guidelines again before saying anything else
19:00:57bluebrotherNico_P: I thought this is the default if not set?
19:01:00linuxstbbarrywardell_: Shouldn't the "lcd_clear_display" happen immediately after show_logo() ? So all the messages are displayed in case of error?
19:01:18Nico_Pbluebrother: I think the default is rather to not touch the files at all
19:01:31jhMikeS#if 0 thread.c line 164 and ata.c line 74
19:01:44bluebrotherhmm. Ok, I had something different in my head. But maybe my memory is wrong ;-)
19:01:54bluebrothermaybe I should check the svn manual ... one day ;-)
19:02:17barrywardell_linuxstb: it happens before the first message is printed
19:02:38barrywardell_so we shouldn't miss any messages
19:02:39linuxstbbarrywardell_: But IIUC, only if the user has forced on the messages. Not if the messages are displayed because of error.
19:02:59n1sbluebrother, Nico_P: if a file has svn:eol-style set to 'native' it will be chaecked out with CRLF on windows and LF on *nix
19:03:13fmlbarrywardell_: is it somehow possible to capture the output of the bootloader? I could then post it for the further analysis.
19:03:37Nico_Pn1s: yes. it's the default behaviour when svn:eolstyle isn't set that we're not sure about
19:03:56*Nico_P is preparing a commit to set svn:eolstyle on the files
19:03:59n1sNico_P: aha, i misread :/
19:04:17linuxstbbarrywardell_: i.e. the case where fatal_error() is called.
19:04:37jhMikeSisn't it "svn:eol-style" ? afaics the property was set on most stuff after the move to CVS.
19:04:51bluebrotherNico_P: thanks. I thought it defaulted to the opposite.
19:05:02jhMikeS*from CVS
19:05:09 Nick d34df00d is now known as rm (n=0xd34df0@
19:05:34Nico_PjhMikeS: you're right, it's svn:eol-style, I mistyped as usual :) it's not set on rbutil files ATM
19:05:37 Join daskhweg [0] (
19:05:46barrywardell_linuxstb: hmmm. true. there's the same problem with the other pp bootloaders. but I didn't notice because there was enough printed to cover the logo completely
19:05:56Nico_Pthankfully SVN's tab completion is very complete
19:06:20soapFWIW, if the rockboxlogo.<W>x<H>x<D>.bmp is larger than expected the build errors out.
19:06:23barrywardell_fml: you could compile your own and add in some sleep(5*HZ) to make it pause for 5 seconds
19:06:50 Nick rm is now known as d34df00d (n=0xd34df0@
19:07:09fmlbarrywardell_: and then? Type it off the screen? Or write to a file? Or make a screen dump?
19:07:59barrywardell_yeah, copy it down from the screen. the sleep will just give you more time
19:08:34barrywardell_soap: yeah, that's known. the bootloader has to be small enough to fit in iram
19:09:12bluebrotherdomonoky: what do you think about removing the "save backup" for non-automated install and placing a checkbox on the dialog?
19:09:34domonokybluebrother: sounds good..
19:09:38bluebrotheralso, the backup could decide (or at least propose) a filename based on the revision string found in rockbox-info.txt
19:09:55domonokygood idea..
19:09:57bluebrotherusers will use arbitrary stuff and only tend to confuse themselves ;-)
19:10:02soaplinuxstb, the ipodpatcher command to install the non-built-in bootloader?
19:10:31bluebrothermy idea is something like backup to /.backups/ and only tell the user where he finds the backup
19:10:38Nico_Pdomonoky, bluebrother: does rbutilqt.rc need to have CRLF (windows) line terminators?
19:10:41bluebrotherwe're writing to the player anyway.
19:11:06bluebrotherNico_P: I don't know for sure, but I think it doesn't make a difference (at least for MinGW)
19:11:08domonokybluebrother: so dont ask for the filename ?
19:11:21bluebrotherdomonoky: yes, that was my idea.
19:12:01bluebrotherwe could do the same for backing up the bootloader on mi4 devices −− just tell the user the backup is placed in /.backup on the player and advice him to copy that to a safe location (like to a CD)
19:12:41soaplinuxstb, if I'm understanding the wiki page right it would just be "ipodpatcher -a bootloader-ipodvideo.ipod"?
19:12:48bluebrotherwhile I don't like rbutil not asking for the bootloader backup as it's working right now I think it would be easier for the build backup
19:13:27bluebrotherwell, we could of course add a (hidden) selector to the dialog that allows putting in the backup filename and put in the proposed filename initially
19:13:54bluebrotherand once the user enables the "create backup of old rockbox installation" unhide that selector.
19:13:54 Quit fml ("CGI:IRC (EOF)")
19:14:20linuxstbsoap: Yes
19:14:23domonokysounds good, i will try it for the normal rockbox install...
19:17:40jhMikeSlinuxstb: what's wrong with the code? you suspect an assembler bug?
19:17:57bluebrotheroh, and did you notice my comment about rbzip.cpp?
19:18:13linuxstbjhMikeS: I'm just confused, as it appears to be giving two different disassemblies for the same bytes (0x1a000001)
19:18:30 Quit hannesd ("Client suicide")
19:19:02linuxstbI guess I should find an ARM reference manual and check for myself...
19:19:02soaphmm, new bootloader seems to fail to load Apple firmware with the hold-switch trick.
19:19:45linuxstbsoap: Does it work if you restore ipodpatcher's internal bootloader?
19:19:49jhMikeS 1a000001 looks like a branch forward by 8 bytes
19:19:56jhMikeSthe 2nd looks correct
19:20:04soaplinuxstb, just one sec.
19:20:07linuxstbjhMikeS: Yes, which seems wrong (based on my original assembler)
19:20:48jhMikeSif should be 1afffffe or something
19:22:27soaplinuxstb, the version 2.0 bootloader internal to ipodpatcher seems to work ok.
19:22:35Nico_Pbluebrother: committed
19:22:46jhMikeSnow with write buffering enabled I get 480fps/1705.5fps RGB and 113fps and 448fps YUV at 264MHz
19:22:48linuxstbsoap: Hmm... It seems easy to upset the Apple firmware.
19:23:23linuxstbsoap: Is this with the bootloader bitmap patch?
19:23:41soapI'm going to try again to make sure I didn't mess up the install (yes this is with the bootloader bitmap patch)
19:23:42n1sjhMikeS: up from what?
19:24:03soapbitmap works fine, error text appears fine
19:24:12linuxstbsoap: It could also be worth testing plain SVN. I have a feeling the bitmap may be overwriting something useful in IRAM that upsets the Apple firmware.
19:25:08soapcan do.
19:27:46*jhMikeS doesn't seem to see a gigabeat F/X test on that framerate page
19:28:43soaplinuxstb, vanilla SVN works fine.
19:29:56soapJust in the name of full disclosure - I did tell rbconf I had a 64MB ipod video, not the default 32 - but that works in SVN, just not with barrywardell_'s patch. (which appears to do all but load Apple firmware)
19:30:08 Join Mathiasdm [0] (
19:30:44 Join webguest99 [0] (n=413cc33c@gateway/web/cgi-irc/
19:30:45soapI need to go run some errands
19:31:02linuxstbsoap: Did you try a 32MB build with barry's patch?
19:31:29webguest99can anyone direct me to where i can get the automatic installer for a 30GB iPod Video??
19:31:47linuxstbwebguest99: It's linked in the install section in the Rockbox manual.
19:32:46webguest99I don't see it in there
19:32:50Buschela fix for FS #5162 is quite easy 1 line of code for iPods and same for e200. is this wanted or will this somehow interfere with sending scrollevents via event? it is already part of the patch FS #8668
19:33:22BigBambiwebguest99: I do:
19:33:23Nico_PjhMikeS: I've disabled the hack and I'm not experiencing the responsiveness issues I used to have without it
19:33:33bluebrotherNico_P: thanks
19:33:39webguest99thank you
19:33:43*barrywardell_ suspects linuxstb is right about the overwriting useful stuff in iram. the patch doesn't do anything else that could be causing problems.
19:34:04BigBambiwebguest99: That is the manual I linked to
19:34:05linuxstbbarrywardell_: I know the OF bootloader puts some stuff at the end of IRAM, we just ignore that fact for now though...
19:35:41jhMikeSNico_P: so you think it's safe to get rid of now? :)
19:35:48bluebrotherhmm, is the tracker broken? I can view tasks but the tasklist gives me only an empty page
19:36:26Nico_PjhMikeS: it looks like it :)
19:37:20Nico_PjhMikeS: does it actually make sense?
19:38:45jhMikeSyes, because mutex use can sleep the processor but the old spinlock thing cannot and the issue came when sleeping the core and issuing ata commands before power has stablized
19:39:23jhMikeSit would just freeze up for a bit in the interrupt wait
19:39:33***Saving seen data "./dancer.seen"
19:39:55*jhMikeS has to run out for a couple hrs
19:40:19barrywardell_bluebrother: it's just the main tracker page at all the rest works still
19:48:36XqtftqxHey guys, for the gigabeat s splash, whats the dimensions?
19:49:15webguest99BigBambi: Thank you.
19:49:40BigBambiwebguest99: no problems :)
19:49:49 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
19:49:55 Quit linuxstb (Nick collision from services.)
19:50:13 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
19:51:17linuxstbjhMikeS: I found my problem - it was caused by me using "arm-elf-objcopy" to extract the binary code. I need to run it through the linker first, then extract the binary code... It then changes the instruction to 0x1afffffd, which is correct.
19:52:52webguest99Anyone know of a site where I can get LEGAL roms from?
19:53:04 Join Synergy6 [0] (
19:53:17scorchewebguest99: google...this isnt the place to ask about ROMs, really
19:53:24scorcheXqtftqx: `look in the code...
19:53:25webguest99Oh okay
19:53:30webguest99My bad
19:55:41webguest99I just installed the base wad for rockdoom, but for someone reason whenever I go and try to run it, it says that i'm missing the base wad
19:56:50webguest99Anyone know what I should do?
19:56:54 Join Yama282 [0] (
19:57:30scorchecheck to make sure that you did it exactly as the instructions tell you?
19:57:51webguest99I did.
19:59:06webguest99I think I figured out what I did wrong.
20:00:00webguest99It's still not working.
20:02:30 Join tvelocity [0] (n=tony@
20:06:39bluebrotherwebguest99: try rbutil −− it can install the base wad for you
20:06:40 Quit ender` (" Everything we know about the Devil has been told us by the the friends of God.")
20:07:17webguest99Bluebrother: Where can I get that from?
20:08:18*linuxstb manages to insert about a 5-second delay loop in his Clip's firmware, without bricking it ;)
20:09:05 Join bluebroth3r [0] (n=dom@rockbox/staff/bluebrother)
20:10:01 Quit barrywardell_ ()
20:10:13 Quit bluebrother (Read error: 104 (Connection reset by peer))
20:11:06Bagderlinuxstb: you rock! that's just great
20:11:07kugellinuxstb: Nice!
20:11:22advcomp2019linuxstb, very nice
20:12:06 Quit csc` (Remote closed the connection)
20:12:07kugelBagder: Would be nice if you commit
20:12:19kugelBagder: Just type and clean up
20:12:42 Join ender` [0] (
20:12:55kugelBagder: Especially the last hunk, I'm not sure how these 8 light got there, or why I forgot to remove them
20:15:54fxbBagder: do have the S5L8700 datasheet?
20:16:33 Join davidc__ [0] (
20:17:00*kugel wonders why the build server aren't updated to use command line options of tools/configure yet!
20:17:36Bagderwhy would it?
20:21:13 Quit davidc__ ()
20:21:37BagderI probably have around 10 scripts doing the same stunt
20:23:22Bagderwhat's the objdump -m switch for big endian arm?
20:23:57domonokybluebroth3r: i commited an improvement to rbutils backup, take a look if you like it more now.. :-)
20:24:46kugelhow's 25s to build a e200 build?
20:25:02 Quit gregzx (Connection timed out)
20:25:02kugeljust installed ccache, major speed up
20:27:02Bagderoh I bet I need to rebuild my tool chain to get that ability
20:27:10kugel1 min w/o ccache
20:28:14n1sBagder: something's wrong with the tracker, when trying to display a list of all tasks it gives an empty page like
20:28:58BagderLinus upgraded the PHP version on the server, I bet that's why
20:29:11kugelWhy didn't I read something about ccache until today? =(
20:29:13webguest99Anyone here familiar with Loader2Args?
20:29:27kugelwebguest99: No, what's that?
20:29:48kugelwebguest99: I just know that it's off-topic if it isn't related to rockbox
20:30:09*linuxstb commits his Sansa hacking experiments to SVN - utils/AMS/hacking/
20:30:17webguest99Its a bootloader that allows you to, when you start up your iPod, to choose what program to start first.
20:30:25bluebroth3rwebguest99: rbutil is linked on the download page ... and loader2 is IPL, ask there.
20:30:34bluebroth3rwe do not support loader2
20:30:37 Join barrywardell [0] (n=barrywar@
20:30:39 Nick bluebroth3r is now known as bluebrother (n=dom@rockbox/staff/bluebrother)
20:30:42webguest99Oh okay
20:30:59Xqtftqxmknboot (rockbox.bin) (orginal.bin) nk.bin
20:31:07Xqtftqxis that correct? or am i mixed up
20:31:25n1sXqtftqx: try it the way suggested on the wikipage...
20:32:56Bagderah for future reference: objdump's option for big-endian is -BE
20:32:57Xqtftqxand rockbox is the boot file?
20:34:10n1sXqtftqx: bootloader.bin is what the bootloader file is named as that page says, when you use different terms than the documentation it is difficult to know what you are asking.
20:34:28XqtftqxOk, sorry, im good now
20:36:43 Quit webguest99 ("CGI:IRC")
20:36:49bluebrotherdomonoky: two issues: there is too much space in the backup group, and you missed inserting a / between mountpoint and backup path (I get "/tmp.backup/<filename>.zip, /tmp being my fake mountpoint)
20:37:03kugelBagder: Did you look at my patch?
20:37:44bluebrotherdomonoky: also, I'd prefer to not show the location until the "Backup" checkbox is enabled.
20:38:10 Quit mackes ("Mackes is out")
20:38:17kugelBagder: It really shouldn't change any functionality, except possible issues which I didn't notice yet
20:38:25 Join mackes [0] (
20:38:32 Part mrpockets ("Leaving")
20:38:43 Quit Yama282 (Read error: 110 (Connection timed out))
20:39:08 Quit mackes (Connection reset by peer)
20:39:09domonokybluebrother: i dont understand what you mean with the too much space, but i can fix the other two issues.. :-)
20:39:33 Join mackes [0] (
20:39:46bluebrotherdomonoky: I meant not enough space ;-)
20:40:03domonokyah, the path is too long ?
20:40:08 Quit mackes (Client Quit)
20:41:20bluebrotherthat too, but its too heigh.
20:41:36 Quit BHSPitLappy (Read error: 110 (Connection timed out))
20:42:41kugellinuxstb: Double posts are bad ;)
20:43:36linuxstbkugel: It's OK, I triple-posted.
20:43:50 Quit atrus ()
20:44:12kugellinuxstb: evil!
20:46:49*Bagder added an objdump wiki page now for the next time I forget its options...
20:47:05 Quit Xqtftqx (" ajax IRC Client")
20:47:43 Join Arathis [0] (
20:48:39Bagderlinuxstb: the size of the first block is stored at offset 0xc so we /should/ be able to change that
20:49:02Bagderbut I guess it might hurt if other stuff depends on its size/placement
20:49:30linuxstbBagder: Yes, that's the "FirmwareSize" field I've been modifying.
20:49:56linuxstbIt's the actual size, not the padded size.
20:50:07 Join petur [0] (
20:50:08Bagderah right
20:50:31linuxstbThe value at 0x08 seems related - it's "padded size / 0x200". But I tried changing that, and it didn't work.
20:51:15linuxstbFrom the description atomicpunk put on the wiki, it seems that the value at 0x08 is also used for the size of other blocks, but I don't understand that...
20:51:17 Join gregzx [0] (
20:51:28Bagderme neither
20:52:23linuxstbBut as long as we can write a bootloader in about 200 bytes, we'll be OK ;)
20:53:45domonokybluebrother: i commited another improvement.. :-)
20:55:19*Bagder got a nano2g bootloader image to disassemble
20:56:04n1sBagder: how did you get it?
20:56:06Bagderit is encrypted/scrambled apart from the initial branch and some other minor stuff
20:59:21bertriklinuxstb: not familiar with this kind of hack yet, but maybe you can write a little program that dumps the bootloader in 200 bytes?
21:00:00linuxstbbertrik: What do you mean?
21:00:48 Join jchord [0] (
21:01:29barrywardelllinuxstb: don't know if it's related, but 0x200 is the block size for most disks
21:02:38bertrikI suppose there's a bootloader in the clip that loads the actual firmware and that one of the problems is that you don't know yet how exactly it loads the main firmware. Once you have a bootloader image, that should be easier to figure out.
21:03:10 Quit jchord ("oops")
21:07:24bluebrotherdomonoky: you're only doing this to improve your commit count ;-)
21:08:15domonokybluebrother: sure, as always commit count and fame is everything.. :-)
21:08:57bluebrotherhehe. Much better now ... but is it (easily) possible to display elides in the backup file name when the space isn't enough?
21:09:15 Quit petur (Remote closed the connection)
21:10:20 Join petur [50] (n=petur@rockbox/developer/petur)
21:11:08 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
21:11:26domonokyhm, on windows the window resizes, so the filename always fits....
21:11:48 Quit BHSPitLappy (Remote closed the connection)
21:12:27bluebrotherwell, I can resize the window here. But it doesn't automatically do this (and I wouldn't like the window resizing because of a single string)
21:14:43domonokywell, i can shorten the filename if it is over X characters with leading dots.. suggest a good X.. :-)
21:14:57BagderGuinness, disassembly and chocolate. what can be better on a warm Sunday evening?
21:15:08soaplinuxstb, do you think it is a worthwhile experiment to build a "32MB" bootloader w/barrywardell's patch?
21:15:43linuxstbsoap: Yes, I think so. All the ipod bootloader builds are 32MB, so maybe something assumes that.
21:15:54soapwill do
21:17:15 Join das-perceptron [0] (
21:17:33bluebrotherdomonoky: well, the table class can automatically add elides if the title is too long. Isn't there something like that available?
21:17:38peturBagder: sitting outside?
21:18:41Bagderno... I gotta be able to hear my son
21:19:09*petur plans to use a webcam for that ;)
21:19:45Bagderyeah, I can in fact use my "baby watcher" but I didn't bother today
21:20:22soaplinuxstb, 32MB build exhibits the same issue, drive never stops running on an attempt to load Apple's firmware.
21:20:30bluebrothergrrr. /me has a strange loop bug
21:23:25 Quit Arathis ("Bye, bye")
21:23:44 Join Xqtftqx [0] (i=40e9e3d2@gateway/web/ajax/
21:24:03linuxstbbertrik: Yes, getting a hold of that code might be helpful, but dumping it would require a disk/usb/something driver, which a) isn't written yet; and b) would be much more than 200 bytes.
21:24:55 Join Arathis [0] (
21:25:34 Part das-perceptron
21:25:52domonokybluebrother: it seems there is no automatic elide for QLabel .. so i would have to take a fixed char count, and shorten myself..
21:26:21bluebrotherdomonoky: wouldn't feeding QString::frameWidth to QFontMetrics::elidedText work?
21:26:32bertriklinuxstb: for hacking powershot cameras the firmware data was blinked out through one of the LEDs on the camera, that might just fit in 200 bytes
21:26:49*domonoky checks..
21:27:16linuxstbbertrik: I don't think we're at the stage for those kinds of hacks yet ;)
21:27:30*petur blinks
21:28:00linuxstbAnyone know how the wheel light works on the E200? Is it just a GPIO?
21:28:00bluebrotheror QString::size and then the width property of the resulting QSize?
21:29:03domonokybluebrother: this function looks promising, but i should you the width of the diplay label, and not of the string :-)
21:29:04bertriklinuxstb: it appears so
21:29:25bertrik_button_light_on has GPIO_SET_BITWISE(GPIOG_OUTPUT_VAL, 0x80)
21:29:48bluebrotherdomonoky: well, I meant using the width of the label itself −− shouldn't that be the width of its display size?
21:29:57*bluebrother notices his typo
21:29:59linuxstbbertrik: Thanks, I just found that as well.
21:30:20domonokybluebrother: :-) s/QString/QLabel and you are right.. :-)
21:31:58 Quit Nico_P (Remote closed the connection)
21:33:44barrywardellsoap, linuxstb: that's an annoying problem with the ipod bl. I don't have an ipod to do any testing with so I'll have to leave this up to someone who has one
21:33:53 Join Synergy66 [0] (
21:35:01bluebrotherdomonoky: seems you can increase your commit count once again?
21:37:04 Quit barrywardell ()
21:38:33domonokyyup, only need to get the correct width..
21:39:34***Saving seen data "./dancer.seen"
21:39:36preglowBuschel: whats with removing the seek table?
21:41:08 Quit Synergy66 ("Adios")
21:50:00 Quit Synergy6 (Read error: 110 (Connection timed out))
21:50:40bluebrotherhmm. Caching for archived builds is broken :(
21:51:04 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:53:01*bluebrother goes back some revisions
21:53:35domonokyit seems also uncached archived builds are broken.. me goes checking..
21:54:05 Quit nplus (Remote closed the connection)
21:54:11 Join n17ikh [0] (
21:56:02 Quit K4rP4D (Read error: 104 (Connection reset by peer))
21:56:22bluebrotherdomonoky: you might hit a broken mirror ... try to use a specific one for testing (I'm using haxx)
22:03:07domonokybluebrother: i cant seem to get the right with for the elide function.. both frameWidth and width are too small (like 5 chars), and maximumSize is too much.. maybe i should just use a fixed size...
22:06:05 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
22:10:00bluebrotherdomonoky: I think I'll give it a try later. Right now I'm struggling with the cache thing :(
22:10:10domonokybluebrother: this cache bug is in installzip..
22:10:48bluebrotherI was looking into that file too ... you already found it?
22:11:02domonokyyep, there a two variables for the cache in installzip, m_usecache and m_path
22:11:22domonokyinstall only sets the path, and there for cache isnt enabled.. :-)
22:11:37bluebrotherI changed install.cpp to call setCache(true) already but it didn't help
22:14:07domonokybluebrother: it works for me, if i use setCache(true) in install.c... (you have to install twice, to fill the cache first)..
22:15:43 Join n17ikh|Lappy [0] (
22:16:22 Quit n17ikh (Nick collision from services.)
22:17:03bluebrotherdoesn't work for me. You changed install.cpp:157?
22:18:23domonokyjup, just changed the "installer->setCache(...) call to use the bool version, and all works nicley..
22:18:54*bluebrother tries a new svn checkout
22:19:26domonokybluebrother: are you sure you didnt disable the cache in the config ? also running it with debug enabled gives good clues if the cache is working..
22:20:30*domonoky increased his commit count again.. :-)
22:20:46bluebrotheryep, I'm sure about the config. Strange ...
22:21:16bluebrotherof course it's possible my recent changes made it act up. But I doubt that.
22:21:53domonokyand we should cleanup installzip.h/cpp it allows to set a cachepath, but doesnt use it..
22:23:37bluebrotherwell, setting a cache path should implicitly enable the cache. I'm about to change this right now
22:24:52bluebrotheror have you included that change already?
22:25:02domonokyyes, as it works for me.. :-)
22:26:30 Join desowin [0] (n=desowin@atheme/member/desowin)
22:27:28Buschelpreglow: there you are :o)
22:27:55Buschelpreglow: the seek table was _never_ used as far as i can see from the code and some experiements
22:33:43 Quit Mathiasdm ("Yuuw!")
22:34:22XqtftqxHey guys, i just installed the updated .rockbox dir and now its stuck on SUM:81DA4AE
22:35:11 Quit m0f0x (Remote closed the connection)
22:37:30 Join m0f0x [0] (
22:37:32XqtftqxNevermind i got it
22:38:42 Quit krazykit_ (Remote closed the connection)
22:40:25 Join saratoga [0] (n=9803c50e@gateway/web/cgi-irc/
22:41:11saratogacan anyone comment on the extra 128k of IRAM on the PP5022/4 ?
22:41:58linuxstb_extra_ 128K?
22:42:16Buschelpreglow: but this is a non satisfying situation −− the seek table now (and before) is allocated and free'ed for each track. int32 array with the size of the number of frames in the track.
22:42:40saratogasorry, addtional IRAM totaling 128k
22:43:08*bertrik checks linker script
22:44:05saratogai am hoping to get around to further optimizing the wma and aac decoders and would like to use that extra IRAM
22:45:12 Quit Arathis ("Bye, bye")
22:48:43n1ssaratoga: the iaudio x5 and probably m5 also have 128k iram
22:49:31linuxstbAre there any targets with 64KB? I know the TCC77x only has that (but it's also armv5, meaning different optimisations)
22:49:51 Quit thegeek_ (Read error: 110 (Connection timed out))
22:50:17saratoganot that i know of, and I think most future targets will have seperate I and D caches making IRAM much less important then it is on PP and Coldfire
22:53:08n1syeah, the caches seem to get significantly better :)
23:01:15Buschelsaratoga/linuxstb: the IRAM configuration is done in
23:01:39linuxstbPartly, and partly in plugin/ (I think).
23:02:25 Quit Zarggg (Read error: 104 (Connection reset by peer))
23:02:52 Join Zarggg [0] (
23:02:55 Join Mathiasdm [0] (
23:03:32Buschelvery interesting fact though. I always thought we use all of the available IRAM for each target
23:05:02 Join jchord [0] (
23:07:16 Join hannesd [0] (
23:08:37 Quit jchord (Client Quit)
23:16:21kugelBagder: So, don't you want to commit it?
23:16:35Bagdernot now, no
23:16:55Bagderbut it looked fine
23:19:30 Quit GodEater_ (Read error: 113 (No route to host))
23:28:22 Join csc` [0] (n=csc@archlinux/user/csc)
23:29:31 Quit Xqtftqx (" ajax IRC Client")
23:29:42peturfor arguments that do not fit on one line, do we align on the previous line indent + 1 tab, or on the ( ?
23:30:26n1spetur: i usually go with the ( and think that's more common in rockbox code
23:30:56 Quit Zarggg ("Pursued by a bear")
23:30:57*petur too, but settings_list is such a mess in that regard
23:32:10peturI think I'll clean it up and hope nobody was too attached to the coding styles in there
23:32:53kugelBagder: If I could, I'd do it myself :)
23:35:07*n1s thins petur should sneak that one into CONTRIBUTING :)
23:36:37n1sI also wonder about the "Language" section in said document, it states that all code should be C, I would like to change that to more picture our current situation...
23:38:26 Quit XavierGr ()
23:39:29linuxstbWhat is our current situation?
23:39:35***Saving seen data "./dancer.seen"
23:39:53n1sasm is ok for time critical stuff
23:42:53 Join cikada [0] (n=472971d2@gateway/web/cgi-irc/
23:43:37cikadahey, does anyone know anything about the insignia mp3 players?
23:43:55 Quit tvelocity ("Αποχώρησε")
23:43:57 Quit cikada (Client Quit)
23:48:29 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
23:53:44 Quit csc` (Remote closed the connection)

Previous day | Next day