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 2011-11-11

00:01:27 Quit ender` (Quit: All great truths begin as blasphemies. George Bernard Shaw)
00:09:40 Quit bertrik (Ping timeout: 258 seconds)
00:35:21 Quit liar (Ping timeout: 258 seconds)
00:42:55 Quit domonoky (Read error: Connection reset by peer)
00:51:13 Quit keyb_gr (Ping timeout: 276 seconds)
00:55:12 Join craig [0] (
00:55:14 Quit craig (Client Quit)
00:55:29 Join craign [0] (
00:58:35 Quit craign (Client Quit)
01:25:09 Quit Thra11_ (Quit: kthxbai)
01:39:54 Quit Jerom (Quit: Leaving.)
01:46:23***Saving seen data "./dancer.seen"
01:59:09 Join ntrly_owl [0] (
01:59:16 Quit advcomp2019_ (Quit: Hi, I'm a quit message virus. Please replace your old line with this line and help me take over the world of IRC.)
02:14:08 Quit hilbert (Quit: Textual IRC Client:
02:19:10CIA-22New commit by fredwbauer (r30962): close file descriptor
02:21:13CIA-22r30962 build result: All green
02:23:39 Quit MethoS- (Read error: Connection reset by peer)
02:26:58 Quit efyx (Remote host closed the connection)
02:40:56 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
02:56:18 Quit Keripo (Quit: Leaving.)
02:57:45LloreanIs there some reason the Rockbox simulator wouldn't be able to play a 4 Go+ file on a filesystem that supports such?
03:01:50 Join fatline [0] (
03:06:07 Join freqmod [0] (~fredrik@2001:700:300:1430:226:18ff:fe82:1a24)
03:06:19freqmodHello, how is the support for clip zip?
03:06:24freqmoddoes it work with the installer?
03:12:27funmanfreqmod: no
03:12:48funmanit's purposedly disabled
03:12:56 Join Keripo [0] (
03:13:10*freqmod was looking into getting one
03:13:21freqmodand thought i would recomend one to a friend
03:13:41freqmodi'll manage to install it anyways
03:13:45funmanit should probably work for daily usage though
03:14:02freqmodhe is happy as long as it plays flac
03:14:33freqmoddoes sd card work?
03:24:22funmanfreqmod: check the 'Sansa Clip Zip' link on the front page
03:24:46freqmodi read the port status
03:26:15freqmodit was not mentioned in the first section
03:28:09 Quit fatline (Quit: This computer has gone to sleep)
03:39:36 Quit Keripo (Quit: Leaving.)
03:46:24***Saving seen data "./dancer.seen"
04:03:25 Quit advcomp2019 (Read error: Connection reset by peer)
04:04:10 Join advcomp2019 [0] (
04:04:10 Quit advcomp2019 (Changing host)
04:04:10 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
04:11:05 Quit amiconn (Disconnected by services)
04:11:06 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:11:28 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:12:08 Quit pixelma (Disconnected by services)
04:12:10 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:12:12 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:26:42 Join fatline [0] (
04:56:57 Quit TheSeven (Disconnected by services)
04:57:09 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
05:02:43 Quit fatline (Quit: This computer has gone to sleep)
05:03:58 Join fatline [0] (
05:06:58 Join [Saint] [0] (~Saint]@
05:06:58 Quit [Saint] (Changing host)
05:06:58 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
05:09:47 Join Rob2223 [0] (
05:13:35 Quit Rob2222 (Ping timeout: 258 seconds)
05:22:05 Quit Llorean (Ping timeout: 240 seconds)
05:23:22 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
05:32:20 Quit Horscht (Quit: Verlassend)
05:33:43 Quit Llorean (Read error: Connection reset by peer)
05:38:19 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
05:41:22 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
05:44:43 Quit [Saint] (Ping timeout: 258 seconds)
05:46:25***Saving seen data "./dancer.seen"
05:48:05 Quit Scromple (Ping timeout: 240 seconds)
06:09:13 Quit Xerion (Read error: Connection reset by peer)
06:09:46 Join Xerion [0] (
06:10:58 Quit zchs (Ping timeout: 252 seconds)
06:19:53 Join zchs [0] (
06:34:34 Quit robin0800 (Quit: Leaving)
06:41:59 Quit froggyman (Quit: Ex-Chat)
06:42:16 Join froggyman [0] (~seth@
06:42:21 Quit [Saint_] (Ping timeout: 255 seconds)
06:42:34 Nick froggyman is now known as Guest54380 (~seth@
07:08:08 Join Topy [0] (
07:10:35 Quit T44 (Ping timeout: 240 seconds)
07:15:04 Quit mc2739 (Ping timeout: 260 seconds)
07:15:22 Join mc2739 [0] (~mc2739@
07:15:32 Quit mc2739 (Changing host)
07:15:32 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
07:24:17 Quit fatline (Quit: This computer has gone to sleep)
07:35:30 Join BHSPitMonkey [0] (
07:35:32 Join fatline [0] (
07:35:32 Quit BHSPitMonkey (Changing host)
07:35:32 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
07:36:54 Join T44 [0] (
07:40:04 Quit Topy (Ping timeout: 240 seconds)
07:44:30 Join Scromple [0] (
07:46:26***Saving seen data "./dancer.seen"
08:02:43 Quit fatline (Quit: This computer has gone to sleep)
08:05:12 Quit Scromple (Quit: Leaving)
08:08:06 Join bertrik [0] (
08:08:06 Quit bertrik (Changing host)
08:08:06 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:16:37 Join GodEater_ [0] (93722cc9@rockbox/staff/GodEater)
08:18:39 Join Keripo [0] (
08:25:25 Join Topy [0] (
08:25:53 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:27:35 Join liar [0] (
08:28:05 Quit T44 (Ping timeout: 260 seconds)
08:32:39 Quit bzed (Ping timeout: 244 seconds)
08:34:46 Join bzed [0] (
08:46:46 Quit amiconn (Remote host closed the connection)
08:46:47 Quit pixelma (Read error: Connection reset by peer)
08:50:33 Join pixelma [0] (quassel@rockbox/staff/pixelma)
08:50:33 Join amiconn [0] (quassel@rockbox/developer/amiconn)
08:52:14 Quit bertrik (Ping timeout: 248 seconds)
08:52:35 Quit factor (Ping timeout: 240 seconds)
08:54:51 Quit BHSPitMonkey (Remote host closed the connection)
08:57:26 Join ender` [0] (
09:25:55 Quit othniel (Ping timeout: 260 seconds)
09:32:52 Join LinusN [0] (
09:39:45 Join mortalis [0] (
09:46:28***Saving seen data "./dancer.seen"
09:52:40 Quit mortalis (Quit: CGI:IRC)
10:10:15 Quit n17ikh (Ping timeout: 260 seconds)
10:10:50 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
10:13:33 Join factor [0] (~factor@
10:16:22 Join n17ikh [0] (
10:26:29 Quit liar (Read error: Operation timed out)
10:37:47 Quit parafin (Remote host closed the connection)
10:38:26 Join parafin [0] (
10:45:51 Quit GodEater_ (Ping timeout: 265 seconds)
10:58:39 Quit bluebrother (Read error: Operation timed out)
10:58:52 Quit fs-bluebot (Ping timeout: 240 seconds)
11:00:32 Join fs-bluebot [0] (
11:00:56 Quit Rob2223 (Quit: Rob2223)
11:00:59 Join bluebrother [0] (
11:01:00 Quit bluebrother (Changing host)
11:01:00 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
11:05:34 Join Rob2222 [0] (
11:20:43CIA-22New commit by theseven (r30963): Fix failed merge of r30953 to 3.10 branch. The ipodnano2g/ipod6g targets should compile again.
11:40:04 Quit Topy (Ping timeout: 240 seconds)
11:46:20 Join Topy [0] (
11:46:31***Saving seen data "./dancer.seen"
11:47:08 Join Stummi|afk [0] (
11:47:08 Quit Stummi (Read error: Connection reset by peer)
11:54:45 Nick Stummi|afk is now known as Stummi (
11:54:45 Quit Stummi (Changing host)
11:54:45 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
12:00:00desowincan there be rename in audio codecs? the tlv320.c seems to cover only tlv320aic23b whilst Sansa Conenct uses tlv320aic3106 which seems to be software incompatible. In Connect's firmware there seems to be support for the both versions (I have no idea if there are actual players with the 23b) and it uses "aic23" and "aic3x" respectively
12:05:24 Join othniel [0] (~othniel@
12:06:14othnielI am investigating a percieved error in the Sansa e200 manual.
12:06:42othnielSection 10.3 Frequency has 3 lines of text.
12:07:03othnielThe first is complete and correct and concise.
12:07:37othnielThe 2nd is inappropriate and inaccurate for the Sansa e200 target.
12:08:59othnielThe 3rd line is missing the first letter and even then contradicts line #1 so it should be removed.
12:09:45othnielIn section 10.1 −− May I suggest a new reading...
12:10:47othnielAfter the word which add "of the four formats" & remove format (singular)
12:12:15 Join Jerom [0] (~jerome@
12:13:40 Join Hendrik_de [0] (~radke@theorie-dyn-29-179.Informatik.Uni-Oldenburg.DE)
12:18:49 Quit kugel (Read error: Connection reset by peer)
12:18:51 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
12:24:54 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
12:27:07 Join hilbert [0] (
12:27:36 Join liar [0] (
12:28:12Torneothniel: raise a bug on the bug tracker; feel free to propose entire new wording, or better still an actual patch to the LaTeX sources that makes the change ;P
12:28:22Torneothniel: but just a note explaining what yout hink is wrong is fine
12:29:34user890104how do i report a bug in the main website, i don't see an suitable category in the bug tracker?
12:31:46Torne..where on the website?
12:32:25user890104shows the changes since 3.8, not since 3.9.1 :)
12:32:44TorneOh, yeah
12:32:48TorneWe know about that, I think :)
12:33:20*Torne pokes it
12:33:44user890104if it's manually at the moment, i can be made to use tools/ i think
12:36:38Hendrik_deHello, Rockboxers. I have a question concerning a certain, MP3 Player, the Medion MD-83377.
12:36:38Hendrik_deI have already opened it and identified the chips, it seems to run with a SigmaTel STMP3770
12:36:38Hendrik_deBig chip on the backside: STMP3770 / A2 PTX / AA0829F / TAIW
12:37:25Hendrik_deOn the front side, below the display, there is some Toshiba chip, probably the display controller: "Toshiba X01758 / Taiwan 08289AE / TC58NVG4D1DTG00"
12:37:51othnielTorne: I did that a month ago in FS 12318 and yesterday I noticed that the whole chapter on recording was missing in the Sansa targets I browsed. Gevaerts fixed it. This comment is just to get it really fixed.
12:37:57Hendrik_deSo, how is support for the STPM3770?
12:38:03Torneothniel: well, raise another bug..
12:38:12Torneuser890104: no, because that's not sufficient information
12:38:23othnielI'm registered
12:38:23Torneuser890104: what it needs to know is the revision number that the release branch was branched from
12:38:34 Join y4n [0] (y4n@unaffiliated/y4ndexx)
12:38:43Torneuser890104: which is nontrivial to discover automatically (possible, but a pain)
12:38:54Tornewe don't release often enough that it's worth automating, i expect :)
12:39:11pamauryHendrik_de: there is code for in the SVN (it's been renamed i.MX 233) since there is an ongoing fuze+ port and there are tools to extract code from the firmware so doing a port is feasible if you have some disassembling/programming knowledge
12:39:58othnielI have several fuze+ DARs
12:40:23othnielGot them from Newegg when they were 37.99
12:40:30othnielor $41.99
12:40:48othnielNow they are $46.99 (refurbished)
12:41:03Torneuser890104: we're releasing 3.10 soon. I'm updating the release checklist n the wiki to include updating that page, which is probably sufficient ;)
12:41:05pamauryHendrik_de: do you have some photos ? In any case it would be nice to put this information on the wiki
12:41:30othnielThe stock firmware records at 44.1k L + 44.1k R channel from FM radio or mic.
12:41:50othnielWait that's mono from the mic −− sorry.
12:41:53user890104Torne: thanks, i know about the upcoming 3.10 and am testing the 3_10 branch on my players
12:49:05 Quit Bagder (Read error: Operation timed out)
12:49:55othnielgevaerts: Please change 10.1 Format to read as follows: "Your recording will be generated in one of 4 supported formats." Then in sentence 2 Instead of "MPEG Layer 3" it should read "MPEG 2 Layer 3"
12:50:11othnielThis is in the e200 manual
12:52:03 Join Bagder [0] (
12:52:03 Quit Bagder (Changing host)
12:52:04 Join Bagder [241] (~daniel@rockbox/developer/bagder)
12:59:17 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:39:27 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
13:42:09gevaertsothniel: please submit a bug report
13:46:34***Saving seen data "./dancer.seen"
13:47:06Hendrik_depamaury: Yes, I have some disasm/programming knowledge. I have some photos, but they are of bad quality.
13:47:32Hendrik_deWhere would one put information about the chips in some player on the wiki?
13:48:45Zagorevery player has their own page
13:50:11Hendrik_deIs there an index page listing all the players?
13:50:20othnielGevaerts: Yessir.
13:50:57ZagorHendrik_de: a bunch are on the wiki front page
13:51:26Zagorthe wiki isn't always quite as clean and sorted as we'd like :-)
13:52:11Hendrik_deYes, saw those, but my player is not in there. Should I just open a new page?
13:52:28 Quit zchs (Ping timeout: 240 seconds)
13:53:40ZagorHendrik_de: what player is it?
13:53:54Hendrik_deIt's a Medion MD-83377
13:54:35 Join Thra11 [0] (
13:55:17Zagoryes, make a new page
13:59:28Hendrik_deHow do I login to the rockbox page? I just registered to edit the wiki.
14:00:29 Quit domonoky (Read error: Connection reset by peer)
14:01:19Hendrik_deCould someone add me (HendrikR) to the WikiUsersGroup?
14:07:56 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
14:10:04 Quit Keripo (Quit: Leaving.)
14:14:37 Quit Topy (Ping timeout: 276 seconds)
14:23:16pamauryHendrik_de: there is some information related to this chip in SansaFuzePlus, SansaFuzePlusPort, Sigmatel, SbFileFormat. You should create a page with the name of your player and put some information and links to it
14:23:26pamauryIf you can take good quality pictures it really helps
14:23:55Hendrik_deI'll see what I can do. But before I can edit the wiki, someone has to add me the the WikiUsersGroup.
14:24:34pamauryZagor: who can edit the wiki ?
14:25:09pamauryHendrik_de: I have to leave, do you have a copy of the firmware / a link to a firmware upgrade so I can see if it's encrypted with the zero key or not ? That will be the major question to answer before starting the port
14:26:11Hendrik_deDon't have the player/software here right now, I'll try to find it when I get home.
14:35:29 Quit FOAD (Ping timeout: 260 seconds)
14:35:46 Join FOAD [0] (~foad@
14:35:46 Quit FOAD (Changing host)
14:35:46 Join FOAD [0] (~foad@unaffiliated/foad)
14:36:02Zagorpamaury: everyone in WikiUsersGroup can edit the wiki, and add others to the WikiUsersGroup page
14:36:30Hendrik_deZagor: could you add me to the WikiUsersGroup, please?
14:36:38Zagorwiki name?
14:45:32Zagorplease register with your full name
14:46:59Hendrik_deOkay, I re-registered myself with full name (HendrikRadke).
14:52:19 Join zchs [0] (
15:03:19 Quit Zarggg (Quit: Rebooting client...)
15:05:07saratogapamaury: sbtoelf does seem to work on that sony player with the zero key
15:06:36saratogaor at least it spits out a bunch of ARM assembly
15:08:56amiconnothniel: MPEG Layer 3 is correct because it can be either MPEG1 Layer 3 or MPEG2 Layer 3
15:12:13 Quit Jerom (Quit: Leaving.)
15:12:22saratogai'm curious about the 4GB limit in rockbox as an app, is that just due to the file pointer being 32 bit?
15:12:28saratogaor is it more complicated
15:14:11Zagorwhat is limited to 4GB?
15:14:53Tornefile sizes
15:15:15amiconn2 GB
15:15:27Tornesaratoga: it's probably two things: using 32-bit offsets, and not calling the 64-bit versions of native file APIs
15:15:43 Join Topy44 [0] (
15:15:43Tornewe can probably fix it iwth a bit of #define'ing
15:16:21Zagor"long fileoffset"
15:17:00saratogaTorne: maybe post that here then:,29266.0.html
15:17:18Tornewe porbably want off_t or fpos_t
15:17:20Torneinstead of long
15:17:32Torne(fpos_t is C99)
15:19:02Tornesaratoga: done
15:19:25amiconnI think it's worth checking how much binsize it would cost going 64 bit for off_t & friends
15:19:47amiconnThe 2 GB filesize limit hits even on non-raaa, as fat32 supports 4 GB file size
15:20:28*amiconn once tested that on H300 - the OF does play a 2.5 GB wav file, rockbox does not
15:20:57saratogaor at least typedef something that can be 64 bit on targets with enough RAM
15:22:23saratogaon a side note, i skimmed through the dircache code ages ago and noticed it does things like use values read from the file system as offsets to pointers, i don't know much about file systems, but perhaps this is part of the reason dircache crashes so easily when parsing corrupted file systems?
15:22:29saratogasome sanity checking might be a good idea
15:23:27Torneamiconn: Well, you can support 4GB files without making off_t 64-bit, if you are sufficiently careful
15:23:28 Join fatline [0] (
15:23:34Torneit requires some somewhat fiddly code
15:23:39Torne64-bit is certainly easier
15:23:47Torneso yeah, i guess check the binsize impact
15:24:00kugel_saratoga: where does it do that?
15:24:34Zagorwhy is unsigned int fiddly?
15:24:52 Quit Topy44 (Ping timeout: 240 seconds)
15:24:57saratogakugel_: let me see if I can remember ...
15:25:02kugel_can't use unsigned as you cant detect errors anymore
15:25:20TorneZagor: because file offsets in general have to be signed
15:25:23Tornebecuse of things like SEEK_CUR
15:25:27Tornei.e. you can seek backwards
15:25:46TorneSo you have to be really careful how you interpret things in different places :)
15:26:38Zagorright, the api breaks but the internal struct could still be unsigned. that would allow reading 4 GB, just not lseeking >2GB
15:26:53Tornethat's what i'm counting as fiddly
15:27:15 Quit Guest54380 (Changing host)
15:27:15 Join Guest54380 [0] (~seth@unaffiliated/froggyman)
15:27:29saratogakugel: just looking again at dircache.c, I see things like "strcpy(ce->d_name, sab.direntry->name);" which I think depend on that file system strings being valid
15:27:31TorneMS used to ahve all these problems in the AVI decoder :)
15:27:32 Nick Guest54380 is now known as froggyman (~seth@unaffiliated/froggyman)
15:27:33amiconnThe internal implementaition can use whatever is suitable, but afaik we're trying to stay roughly posix compatible
15:27:36saratoga(but i could be wrong)
15:27:42saratogaanyway i have to run
15:28:07TorneRight, yeah. We could abandon C/POSIX compatibility and just have seperate seek forwards/backwards calls ;p
15:28:07Zagorstill, lseek can use 64-bit without the internal having to be
15:28:15TorneOr that
15:28:38Zagorthough I'm not terribly worried that going to 64 bit is going to cost much size or performance
15:28:41TorneIf the binsize impact is negligible, though, it's easier to just define offsets as int64_t
15:29:02 Join Topy44 [0] (
15:37:18 Quit saratoga (Ping timeout: 265 seconds)
15:40:58 Join bertrik [0] (
15:40:58 Quit bertrik (Changing host)
15:40:58 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
15:42:34 Part Hendrik_de
15:45:14 Join MethoS- [0] (~clemens@
15:46:37***Saving seen data "./dancer.seen"
16:04:53 Part LinusN
16:10:12pamaurysaratoga: sbtoelf tells you if you have the right key or not, if it produces the elf files, the encryption key is right
16:16:42 Quit fatline (Quit: This computer has gone to sleep)
16:18:04 Join fatline [0] (
16:23:30 Quit fatline (Quit: This computer has gone to sleep)
16:24:14 Join n1s [0] (
16:24:14 Quit n1s (Changing host)
16:24:14 Join n1s [0] (~n1s@rockbox/developer/n1s)
16:28:59 Part Zagor
16:31:59 Quit liar (Remote host closed the connection)
16:39:12 Join liar [0] (
16:50:16 Join chattr [0] (
16:50:27 Part chattr ("gone")
16:51:58 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
17:04:26 Quit Lucifer (Read error: Connection reset by peer)
17:04:49 Join Lucifer [0] (
17:22:25 Join WalkGood [0] (~4@unaffiliated/walkgood)
17:46:39***Saving seen data "./dancer.seen"
17:52:15 Join low_light [0] (
17:54:24CIA-22New commit by thomasjfox (r30964): Fix copy'n'paste bug in skin engine ...
17:55:39thomasjfoxJdGordon: Please have a look at this: [firmware/drivers/rtc/rtc_rx5x348ab.c:62] -> [firmware/drivers/rtc/rtc_rx5x348ab.c:62]: (style) Same expression on both sides of '|'.
17:56:17CIA-22r30964 build result: All green
17:59:46thomasjfoxNote sure how to fix this yet: [firmware/drivers/rtc/rtc_zenvisionm.c:38]: (error) Returning pointer to local array variable
18:00:03thomasjfoxMaybe we should make the array "static"?
18:02:44gevaertshm, is that function ever used?
18:03:08gevaertsrtc_read() is
18:04:15gevaertsnot with that prototype
18:06:15funmanthe file is not built anyway
18:06:26thomasjfoxSo should we just remove it?
18:06:42funmanthomasjfox: are you making a zenvisionm build?
18:06:58thomasjfoxNope, just checking the code base in general
18:07:15thomasjfoxBut it's part of our auto-build
18:07:18funmanfix it properly or leave it as-is for the next person who'll work on this port
18:07:52 Join ChickeNES-laptop [0] (~ChickeNES@
18:08:18thomasjfoxOh well, in that case I'll just leave a comment in the code that it's returning the address of an array on the stack
18:08:22 Quit liar (Read error: Connection reset by peer)
18:08:55thomasjfoxOr make it "static"
18:09:03thomasjfoxThat's the better solution
18:09:39funmanyou could use the correct prototype also
18:10:19funmanaah no sorry
18:10:33funmanrtc_read is just a kind of alias to i2c_read for rtc
18:10:41funmani was confused with rtc_read_datetime
18:10:44*gevaerts isn't sure if cleaning up code in an unused file of an inactive port is really worth much time :)
18:11:37thomasjfoxLike funman said, if someone continues at that place, strange things might happen if he doesn't catch this subtle error
18:12:49funmanbut someone will get thousands of bigger problems before that so i guess that's okay, and the warning is pretty explicit
18:14:40bertrikI'd rather not change something in a port I don't have
18:14:44 Join liar [0] (
18:14:59funmanthere was something major blocking zen vision port
18:15:38bertrikbut not so long ago, pamaury and I spotted several problem in one of the usb drivers
18:17:29funmanit's a pity because those have a pretty good speaker
18:18:05thomasjfoxOk, one more before amiconn starts barking about static analysis: [apps/codecs/libm4a/demux.c:247]: (style) Same expression on both sides of '-'
18:18:24thomasjfoxThe code is valid, it just looks suspicious
18:18:52thomasjfoxEspecially as there's an "if (entry_remaining)" just some lines below
18:19:40gevaertsDoesn't that one fall under "don't touch imported code without a really good reason"?
18:20:20thomasjfoxI think I got the meaning of it: It should be "size_remaining -= entry_remaining"
18:20:39thomasjfoxWell, once imported, how do we get bugfixes from upstream?
18:21:07thomasjfoxI certainly won't touch it, just wanted to point it out
18:23:51gevaertsI agree that someone should have a good look. It doesn't look right indeed...
18:28:56 Quit Stummi (Read error: Connection reset by peer)
18:28:58 Join Stummi|afk [0] (
18:29:38thomasjfoxbuschel might be the right -someone-
18:40:04 Join TheLemonMan [0] (
18:44:51*thomasjfox ducks
18:44:58CIA-22New commit by thomasjfox (r30965): Fix file descriptor leak ...
18:46:32CIA-22r30965 build result: All green
18:46:36 Join ReimuHak_ [0] (
18:56:43 Quit hilbert (Ping timeout: 276 seconds)
19:01:47 Quit ChickeNES-laptop (Read error: Connection reset by peer)
19:01:47 Join ChickeNE_ [0] (~ChickeNES@
19:03:39 Join hilbert [0] (
19:03:57 Quit hilbert (Remote host closed the connection)
19:06:40 Join AlexP_mob [0] (~AlexP@rockbox/staff/AlexP)
19:13:30 Join Horscht [0] (
19:13:30 Quit Horscht (Changing host)
19:13:30 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:17:57 Quit TheLemonMan (Quit: WeeChat 0.3.6)
19:20:24 Quit ChickeNE_ (Quit: Computer has gone to sleep.)
19:24:36 Quit Horscht (Quit: Verlassend)
19:27:52 Join Horscht [0] (
19:27:52 Quit Horscht (Changing host)
19:27:52 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:29:23 Quit Thra11 (Ping timeout: 240 seconds)
19:29:36 Quit MethoS- (Remote host closed the connection)
19:30:25 Join T44 [0] (
19:31:36 Quit Topy44 (Ping timeout: 245 seconds)
19:37:11 Quit thomasjfox (Remote host closed the connection)
19:46:42***Saving seen data "./dancer.seen"
19:52:05 Quit liar (Remote host closed the connection)
20:02:23 Join Buschel [0] (
20:05:13CIA-22New commit by kugel (r30966): Document the internal alpha channel format better, and fixes for 32bit alpha bitmaps. ...
20:07:01CIA-22r30966 build result: All green
20:08:54Buschelwill anybody speak up against the latest patch in FS #8668? this will implement cpu boosting on scrolling. the boost will end after timeout. I can only test for nano2g and iPod Video and will therefor only activate this for 4G scroll wheels
20:08:55fs-bluebot battery runtime: experimental gui boost (patches, assigned)
20:10:27 Quit Stummi|afk (Read error: Connection reset by peer)
20:10:28 Join Stummi [0] (
20:10:30 Quit Stummi (Changing host)
20:10:30 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
20:13:42 Join saratoga [0] (9803ec71@gateway/web/freenode/ip.
20:13:48 Nick kugel_ is now known as kugel (~kugel@rockbox/developer/kugel)
20:13:50saratogaBuschel: no, in fact I have been strongly in favor for many years now
20:14:16kugelBuschel: it implements it for targets which don't obviously need it, and ignores other targets :)
20:14:24kugeladd e200 :)
20:14:31 Quit sinthetek (Read error: Operation timed out)
20:14:56kugelBuschel: but seriously, IMO it doesn't go far enough. Why limit to scroll wheel actions?
20:15:33saratogayou are referring to the ipod6g?
20:15:39Buschelfrom what I undertand it is not. it is detecting BUTTON_SCROLL_BACK/_FWD
20:15:40kugelnano2g too
20:15:54kugelBuschel: that's scroll wheel
20:16:25kugelBACK/FWD indicate wheed movement. for touchpads there's UP/DOWN
20:16:31AlexP_mobDo non scroll wheel targets not benefit?
20:16:52kugelAlexP_mob: only select ipods
20:17:09saratogathey do, but probably a lot less since a button is less latency sensitive then a wheel
20:17:25saratogai have no opinion on the newer Ipods, but I think just implementing for wheels is a good first step
20:17:31AlexP_mobkugel: sorry, I meant would they not
20:17:55saratogapeople can experiment with other devices after wards
20:18:08Buschelyep, exactly my thoughts
20:18:19Buschelif it is in it can be easily adapted
20:18:43kugelit could have been implement this way ages ago. it's unfortunate it took so long
20:19:19Buschelshould I add this for all iPod's (and e200)?
20:19:30bertrikwhat is the problem exactly? and is boosting the only solution?
20:19:47 Quit WalkGood (Quit: me fui)
20:20:14Buschelthe only "problem" is experienced on the iPod Video −− the gui ist lasgy and not responsive
20:20:23saratogabertrik: we'd like to have the normal clock as low as possible to save power, but doing so causes the UI to be less responsive
20:20:45Buschelboosting the CPU while scrolling makes the gui far more responsive
20:20:59kugelespecially with sbs, anti-aliased fonts and alpha bitmaps the UI can be very heavy. some targets need boosting to ensure responsiveness
20:21:03saratogai originally wanted this so that we could lower the PP clock to 24 MHz (or perhaps even lower) and save even more battery life
20:21:43saratogaour codecs are becoming well enough optimized that they often use much less cpu cycles then the UI
20:21:56 Join stoffel [0] (
20:26:03Buschelok, now I have the following devices with gui boost: nano2g, video, sansaview, 6g, 3g, nano1g, e300v2, color, mini2g, fuzev2, 4g, e200, fuze, mini1g, 1g2g. anybody missing any device?
20:26:12bertrikok, but casual inspection shows me that a lot of (probably unneeded) full screen updates are taking place, boosting can make that better, but isn't fixing the root issue
20:26:43bertrik(just an example)
20:27:25Buschelwell, we were seaching for other optimization since a while. boosting the gui on action seems reasonable.
20:28:10Buscheland as saratoga mentioned this is a big step towards lowered default clock which can bring several hours of battery runtime (speaking of iPod Video)
20:28:59saratogabertrik: i'm all for optimization, but realistically ideal clock (from a battery POV) on a lot of targets is <20MHz
20:29:12saratogaon the FuzeV2 we have to keep the PCLK at 40MHz just to keep the LCD feed
20:29:31 Quit factor (Ping timeout: 245 seconds)
20:30:28 Join sinthetek [0] (
20:30:29 Quit sinthetek (Changing host)
20:30:29 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
20:40:34CIA-22New commit by buschel (r30967): Finally submit GUI boost (FS #8668). With this change the CPU is boosted (with a 1 second timeout) on scrollwheel activity in the list, main menu, tree ...
20:42:21CIA-22r30967 build result: All green
20:43:50kugelbertrik: this is only for lists, where only 1 fullscreen update is done
20:44:08 Join Yollean [0] (
20:44:14 Quit Yollean (Client Quit)
20:44:59 Join factor [0] (~factor@
20:48:20 Quit AlexP_mob (Ping timeout: 255 seconds)
20:48:54 Quit GodEater (Ping timeout: 244 seconds)
20:55:22amiconnBuschel, saratoga: GUI boost may be useful on most (all?) colour targets, but then there's still the boost latency problem on several SoCs
20:55:50amiconnOn greyscale and monochrome targets I'd say it's unnecessary
20:57:22Buschelbut it might allow to lower the default clock on such targets
20:57:56amiconnYou could lower that even without gui boost
20:58:28amiconnThere were experiments on H1x0 to lower the default clock to 34 MHz (from 45)
20:58:56amiconnThe gui worked fine - but otoh the gui is slowish on H300 even at 45 MHz
21:03:26 Join dreamlayers [0] (
21:04:05 Quit dreamlayers (Changing host)
21:04:06 Join dreamlayers [0] (~dreamlaye@rockbox/developer/dreamlayers)
21:07:42dreamlayersHow can we be sure that the GUI boost code will unboost when there can be a blocking button_get(true) before it?
21:09:54 Quit factor (Read error: Connection reset by peer)
21:12:04 Quit dreamlayers (Ping timeout: 240 seconds)
21:21:28 Quit sinthetek (Ping timeout: 256 seconds)
21:27:39 Join factor [0] (~factor@
21:31:20*mc2739 wonders if kugel has seen FS #12380 yet
21:31:20fs-bluebot e200v2 - mkamsboot cannot patch bootloader since r30937 (bugs, new)
21:32:10funmando we really care since we have a fully working bootloader already?
21:32:59mc2739no, as long as we never need another update to it
21:34:40 Join scanf [0] (~x32@unaffiliated/scanf)
21:34:58scanfis there a rockbox skin that emulates the original iPod interface/keybindings
21:35:09scanfim not a fan of the default rockbox keybindings (wheel bindings?)
21:35:10kugelmc2739: did the code size increase the bootloader or is there some static array that got larger?
21:35:16 Quit T44 (Ping timeout: 240 seconds)
21:37:02mc2739I haven't looked into what caused the increase, I just bisected to find when the increase occurred
21:37:11 Join T44 [0] (
21:37:32 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243])
21:38:30 Join keyb_gr [0] (
21:43:17scanfcan i determine the decibel level my headphones are outputting from rockbox firmware?
21:46:25gevaertsscanf: rockbox can only show the level compared to the maximum output level. There's no way of getting at the actual dBA level of your headphones
21:46:44***Saving seen data "./dancer.seen"
21:50:29 Join Stummi|afk [0] (
21:50:29 Quit Stummi (Read error: Connection reset by peer)
21:52:09saratogascanf: thats something that has to be measured, not something that software can calculate
22:06:14 Quit froggyman (Read error: Operation timed out)
22:19:42 Join froggyman [0] (~seth@
22:19:43 Quit froggyman (Changing host)
22:19:43 Join froggyman [0] (~seth@unaffiliated/froggyman)
22:21:58 Quit Stummi|afk (Quit: Bye!)
22:24:29 Quit froggyman (Client Quit)
22:28:59 Join Codefree [0] (
22:30:05 Part Codefree
22:31:48 Quit saratoga (Quit: Page closed)
22:32:26 Part low_light
22:46:21 Join TheLemonMan [0] (
22:55:24 Join pamaury_ [0] (
22:55:55 Quit zchs (Ping timeout: 244 seconds)
22:55:58 Quit pamaury (Ping timeout: 260 seconds)
23:03:30CIA-22New commit by kugel (r30968): Handle 32bit bitmaps with all-zero alpha channel as fully opaque. ...
23:05:12CIA-22r30968 build result: All green
23:06:27 Join zchs [0] (
23:07:13 Quit stoffel (Remote host closed the connection)
23:09:02 Join fatline [0] (
23:09:28 Quit keyb_gr (Ping timeout: 252 seconds)
23:13:37 Join keyb_gr [0] (
23:16:10 Join Keripo [0] (
23:33:09 Join metaphys [0] (
23:34:59 Join Topy44 [0] (
23:35:26 Quit TheLemonMan (Quit: WeeChat 0.3.6)
23:37:16 Quit T44 (Ping timeout: 240 seconds)
23:37:46 Quit zchs (Ping timeout: 244 seconds)
23:44:05 Join zchs [0] (
23:46:48***Saving seen data "./dancer.seen"
23:50:17 Join jaykay [0] (
23:52:10jaykayi'm not a C expert, but the gui boost seems to do the unboost-check only when a button is pressed. is this wrong/intended/...?
23:53:14n1sjaykay: have you seen this behaviour or are you inferring it from reading the code?
23:53:38jaykayn1s: from reading the code
23:54:15jaykaycan't check it, would have done that otherwise :)
23:55:42 Nick pamaury_ is now known as pamaury (
23:55:50 Quit pamaury (Changing host)
23:55:50 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:57:57n1sseems to me that it will unboost in get_action_worker if a scroll "button" is not pressed and the timeout has passed
23:58:53 Quit ReimuHak_ (Ping timeout: 248 seconds)

Previous day | Next day