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 2005-01-02

00:07:00 Join jyp_ [0] (
00:07:17 Join gromit^ [0] (
00:08:12 Join jyp__ [0] (
00:26:12 Quit jyp_ (Read error: 110 (Connection timed out))
00:28:14***Saving seen data "./dancer.seen"
01:08:08 Quit jyp__ ("Leaving")
01:28:27 Join elinenbe_ [0] (
01:29:19 Quit elinenbe (Read error: 104 (Connection reset by peer))
01:29:19 Nick elinenbe_ is now known as elinenbe (
01:39:35 Join elinenbe_ [0] (
01:39:37 Quit elinenbe (Read error: 104 (Connection reset by peer))
01:39:41 Nick elinenbe_ is now known as elinenbe (
01:50:22 Join gromit`` [0] (
02:00:00 Quit gromit`` ("Leaving")
02:07:58 Quit gromit^ (Read error: 110 (Connection timed out))
02:28:18***Saving seen data "./dancer.seen"
02:48:40 Quit Stryke` (Read error: 110 (Connection timed out))
02:51:40 Join gromit^ [0] (
04:28:19***Saving seen data "./dancer.seen"
05:15:44 Join skiguy411 [0] (
05:34:15 Quit skiguy411 ()
06:03:46 Quit gromit` (Remote closed the connection)
06:04:29 Quit gromit^ (Remote closed the connection)
06:28:23***Saving seen data "./dancer.seen"
06:38:56 Quit adi|home (Read error: 110 (Connection timed out))
08:28:25***Saving seen data "./dancer.seen"
08:42:17 Join midk [0] (
08:48:41 Join ashridah [0] (
08:49:31ashridahi just had a thought. would people say that music is usually mixed for speakers? would it be possible to rig up rockbox on an iriver (eventually) to cross-mix the audio across channels a bit to stop sound appearing to be inside the skull?
09:09:51 Join methangas [0] (
09:30:33 Join IRCMonkey [0] (
09:30:58IRCMonkeyHello, best wishes for new year to all Rbx members and supporters... :)
09:33:07IRCMonkeyI'm Tang
09:33:24IRCMonkeydon't retrieve how to be recongnised as
09:34:06 Quit IRCMonkey (Client Quit)
09:34:52 Join Tang [0] (
09:35:00Tangok that's fine
09:37:03methangaswell hello there
09:56:47 Quit Nibbler (Read error: 104 (Connection reset by peer))
10:06:55 Join Nibbler [0] (
10:28:28***Saving seen data "./dancer.seen"
10:31:13 Quit Nibbler (Read error: 104 (Connection reset by peer))
10:34:44 Quit midk (Read error: 104 (Connection reset by peer))
10:35:06 Join Nibbler [0] (
11:01:04 Join Bagder [0] (
11:01:10 Join NibbIer [0] (
11:09:37 Quit Nibbler (Read error: 60 (Operation timed out))
11:09:37 Nick NibbIer is now known as Nibbler (
12:24:36 Join [IDC]Dragon [0] (
12:25:39[IDC]DragonHappy new year!
12:28:29***Saving seen data "./dancer.seen"
12:58:52 Join amiconn [0] (
12:59:31amiconnHi everybody, and happy new year!
13:34:25 Join huz [0] (
13:48:25 Join JJ-Demon [0] (
13:50:53amiconn[IDC]Dragon: When you implemented FAT16 support, did you test whether all fat.c functions work as intended?
13:51:32amiconnI wonder how e.g. free_direntries() should work with FAT16 root dir entries. It (unconditionally) uses fat_open()...
13:53:30 Quit huz ("Leaving")
13:56:42 Quit [IDC]Dragon ("CGI:IRC (Ping timeout)")
14:06:57 Join [IDC]Dragon [0] (
14:07:24amiconnHi again
14:15:57[IDC]DragonWLAN of my new notebook sucks
14:17:38amiconnI also had some stability problems with WLAN some time ago.
14:18:23amiconnAlthough it got much more stable after updating the drivers, I don't like WLAN much.
14:20:43[IDC]Dragonmy webpad has no such problems
14:21:09[IDC]Dragonbut it runs linux...
14:22:03amiconnI try to understand the fat driver...
14:24:09[IDC]Dragoncan I help?
14:24:26[IDC]Dragonbut I don't understand it either ;-)
14:25:21amiconnI would like to get rid of the separate fat_open_root() for FAT16, and add the functionality to fat_open().
14:25:27amiconnThe only "extra" would be a little function to calculate the root dir sector offset
14:26:05amiconnI have to check whether this may break something, or where the negative-cluster-check has to be added
14:28:08[IDC]Dragonnegative-cluster-check where?
14:28:33***Saving seen data "./dancer.seen"
14:29:19amiconnnegative cluster check to decide whether adding the root dir sector offset is needed or not. This will be needed within fat_open() and fat_seek() for sure, but there may be more places...
14:32:40 Join zeekoe [0] (
14:33:44[IDC]Dragonhmm, I think open and seek are enough
14:35:01 Quit ashridah ("bbs")
14:41:54amiconn[IDC]Dragon: If you would change HAVE_MULTIVOLUME to a numeric option (e.g. CONFIG_NUM_VOLUMES), it would be very easy to change the number of supported volumes...
14:42:58zeekoehey... can i interrupt a second? :)
14:43:26amiconnzeekoe: You already did ;-)
14:43:31zeekoeah, good
14:43:39zeekoei want to change the screen "Resume? PLAY = Yes Any other = No"
14:43:49zeekoebut i can't find where it's displayed :-/
14:44:12zeekoethe string is in lang.c
14:44:26amiconnIt is located in tree.c. The string itself is localized
14:44:53*zeekoe searches
14:45:22[IDC]Dragonamiconn: CONFIG_NUM_VOLUMES, yes, possible
14:45:40zeekoeha! LANG_RESUME_ASK, that's it
14:45:42zeekoethanks a lot
14:45:56[IDC]Dragonif =1, then like no HAVE_MULTIVOLUME
14:46:33[IDC]Dragondo you think we'd like more than 2?
14:46:37amiconnor simply not defining the symbol
14:47:49[IDC]Dragonthat's complex to query, not exist or ==1
14:48:24 Quit Tang ("Chatzilla 0.9.66 [Mozilla rv:1.7.5/20041107]")
14:48:58 Quit zeekoe ()
14:49:23[IDC]DragonI was briefly looking into MMC hotswap, but quickly dropped it
14:49:45amiconnMy sister just got an error with multivolume :(
14:49:47[IDC]DragonI'm not sure about which ends would need maintenace for that
14:50:10[IDC]Dragonok, let's maintain that first...
14:50:15amiconnShe was listening to music (from MMC) and playing a game in parallel
14:51:17[IDC]DragonI forgot to post a warning
14:51:22amiconnThe plugin shut down (unclean? Counting sokoban up above level 100...). Then the content of the internal flash became unaccessible
14:51:35amiconnMMC still accessible though...
14:51:36[IDC]Dragonuh oh
14:53:35amiconnIt's reproducable...
14:56:12[IDC]Dragonlistening to music is unlikely to attempt writing to MMC
14:56:37[IDC]Dragonso it doesn't sound like a volume mixup to me
14:58:24[IDC]Dragonmaybe the fat cache went bezerk
15:00:56amiconnIt's not reproducable on the Ondio SP - strange
15:03:01[IDC]Dragonperhaps you have a different formatting
15:07:07amiconnNow I didn't manage to reproduce it again...
15:14:20amiconnIt does not happen everytime. I did the follwoing:
15:14:46amiconn(1) Start Ondio with MMC inserted, and start playing some music from MMC.
15:15:13 Quit elinenbe (Read error: 104 (Connection reset by peer))
15:15:29 Join elinenbe [0] (
15:15:33amiconn(2) Go to the menu, and start a plugin (I used sokoban, but this may happen with others as well. The sokoban shutdown after level 100 is normal.)
15:16:19amiconnAfter leaving the plugin, it may happen that you can't start another plugin, but the effect varies.
15:16:41amiconnEither this is temporary, and trying once more works, or not.
15:17:42 Quit elinenbe (Read error: 104 (Connection reset by peer))
15:17:58amiconnIn the latter case the whole internal flash contents is no longer displayed (and rockbox complains when trying to access the playlist control file etc)
15:18:03 Join elinenbe [0] (
15:23:38[IDC]Dragonand then I need to reformat?
15:23:55amiconnNo, a simple reboot is sufficient.
15:24:12[IDC]Dragonnow that is strange
15:24:28[IDC]DragonI suspected it to mess up the file system
15:25:25amiconnchkdsk finds no errors, both internal and MMC
15:26:07[IDC]Dragonit really sounds lik the fat caching
15:28:26amiconnI tried with the SP somre more, and I managed to get the temporary effect too. Sometimes it was unable to load the plugin, or sokoban complained about being unable to load the levels file
15:29:26 Join huz [0] (
15:29:43huzhello guys - I'm really stuck w/ a "hd error" problem
15:29:50 Quit huz (Client Quit)
15:30:03[IDC]DragonI'll try later - no time now
15:30:46 Join huz [0] (
15:30:58huzhello guys - I'm really stuck w/ a "hd error" problem
15:31:24huzthe incriminated HD seems OK - I can mount it w/ windows and Partition Magic finds no error on it
15:31:53huzI've downloaded on the latest rockbox version as well without problem
15:32:04huzHas someone a hint to help me ?
15:34:04huzMy archos is a jukebox 6000 on which I changed the hd
15:37:51amiconnhuz: (1) What error do you get exactly? (2) Does the HD mount correctly from USB, i.e. from within the jukebox, or did you test with an external enclosure or a 2.5" ide adapter?
15:38:34huzamiconn: (1) HD ERROR (2) I mount it direclty from USB
15:40:02amiconnHmm. No more specific message? I guess this message is generated by the archos firmware. What happens if you start with the charger connected? Same thing?
15:40:53huzamiconn : yeah that's from the archos firmware - I boot it with the charger plugged in indeed to avoir some mischarged batteries errors
15:41:09amiconnUsually HD error suggests that there are battery problems (totally discharged, bad battery cell, or broken contacts). However, this usually also means that USB access doesn't work as well
15:42:17amiconnDid it work before, with this very HD? I'm asking because you told that you changed it.
15:44:24huzhmm no I never make it worked w/ this HD
15:44:38huzan i've no other one to make some tests
15:44:44huzthe previous one seems dead
15:45:01amiconnMaybe this hd is not jukebox compatible, or it is jumpered wrong.
15:46:10huzit's a toshiba hd - pretty good one
15:46:27huzOK i'll buy an ipod next time ;)
15:47:24 Join gromit` [0] (
15:47:30amiconnHmm, there were a number of problems reported on the ml over time with toshiba HDs, especially OEM versions seem to cause trouble
15:47:33 Join gromit^ [0] (
15:48:12amiconnIirc this is not unsolvable, maybe some jumpers need to be changed
15:48:42amiconngotta leave now
15:48:58 Part amiconn
15:49:08huzthx amiconn
15:52:05JJ-Demonis anyone in #linav ever alive?
15:54:12 Quit huz ("Client exiting")
16:05:37 Quit [IDC]Dragon ("CGI:IRC (Ping timeout)")
16:18:35 Nick JJ-Demon is now known as JJ-Demon|sleep (
16:18:38 Quit JJ-Demon|sleep ("—I-n-v-i-s-i-o-n— 2.0 Build 3515")
16:27:37 Quit gromit` ("Leaving")
16:27:46 Quit gromit^ ("Leaving")
16:28:36***Saving seen data "./dancer.seen"
16:32:28 Join quelsaruk [0] (~kvirc@
16:50:21 Join jyp_ [0] (
17:31:23 Join amiconn [0] (
17:37:43 Join z35 [0] (
18:03:53 Part amiconn
18:08:54 Join zeekoe [0] (
18:28:38 Join amiconn [0] (
18:28:40***Saving seen data "./dancer.seen"
18:29:35quelsarukhi amiconn
18:33:40quelsarukamiconn: how's the howto going? :D
18:34:05amiconnoh no ;-)
18:38:18zeekoehm... anyone know what the path to .rockbox is, on the box itself? is it just "\.rockbox\" or "/.rockbox/" ?
18:38:50amiconnIt's /.rockbox/ because rockbox uses unix path syntax
18:40:05zeekoei like unix path better than ms path
18:58:53 Join zeekoe2 [0] (
18:59:01 Quit zeekoe (Nick collision from services.)
18:59:10 Nick zeekoe2 is now known as zeekoe (
19:21:56zeekoegrr... is down?
19:22:29quelsaruklet me check
19:23:13quelsaruksourceforge is up
19:24:32zeekoehmyes, the moment i said it, it worked again...
19:25:26 Quit methangas (" Try HydraIRC -> <-")
19:36:34 Join [IDC]Dragon [0] (
19:36:43quelsarukhi [IDC]Dragon
19:36:52[IDC]DragonHi there
19:37:24amiconnhi again
19:38:00 Quit zeekoe ()
19:38:26[IDC]Dragonamiconn: I'm looking at the fat caching code
19:39:09[IDC]Dragonand wonder how this ever has worked
19:39:46[IDC]Dragonour file access isn't atomic, just the sector access itself, right?
19:40:18amiconnI believe so
19:41:00[IDC]Dragoncache_fat_sector() returns a pointer to a fat sector, and may trash others
19:41:13 Join zeekoe [0] (
19:41:34amiconnI still try to understand the fat driver. Most likely I'll clean up cluster2sec() and sec2cluster(), and use it in more places than now
19:41:57zeekoewhee, amiconn, look what i've done with your help :)
19:42:01[IDC]Dragonwhat't wrong there?
19:43:12amiconnThey are not symmetric, and more complicated than necessary
19:43:57amiconncluster2sec() works for both data area and root dir (FAT16), while sec2cluster() does only work for the data area.
19:45:09amiconnFurthermore, cluster2sec() uses a function first_sector_of_cluster(), which is *only* used from there. Both functions are quite small, so they can be unified.
19:45:54[IDC]Dragonmaybe the compiler auto-inlines it already
19:46:59amiconnYes. Still it does look more complicated than necessary
19:51:34amiconn[IDC]Dragon: What's the problem with cache_fat_sector() ?
19:52:13[IDC]DragonI'm not sure if the pointer will be used across a context switch
19:52:47[IDC]Dragoncaching one sector may thrash another, being used in another thread
19:56:48amiconnHmm. I don't think these pointers are used across context switches, although I'm not sure
19:58:17amiconnGotta go for a while
19:58:24 Part amiconn
19:59:03*[IDC]Dragon is away, too
20:06:50 Quit zeekoe (Read error: 60 (Operation timed out))
20:28:42***Saving seen data "./dancer.seen"
21:05:22 Join lImbus [0] (
21:06:32lImbusgood evening everybody
21:25:41 Quit [IDC]Dragon (Read error: 110 (Connection timed out))
22:00:55 Join gromit` [0] (~gromit`
22:28:46***Saving seen data "./dancer.seen"
22:31:26 Join z351 [0] (
22:40:32 Join bagawk [0] (
22:40:40 Join amiconn [0] (
22:42:54bagawkHey amiconn
22:43:04bagawkHej Bagder
22:43:41Bagderhi ho
22:47:59 Quit z35 (Read error: 110 (Connection timed out))
22:49:32bagawkMy dad finaly relized how horrible AOL is after he got a trojan horse dialer
22:50:21bagawka nice 256k DSL will be installed soon
22:56:46lImbusWTF ? 256k dsl ? is that the upload or the price ?
22:57:04lImbusor the amount of average clients
22:58:02lImbusbelgium has 3300/786 kBits dsl per default
23:01:35bagawk256k up and down
23:01:42bagawki know kinda slow
23:01:56bagawkBut hey, atleast I got him away from dialup/AOL
23:02:14lImbusyeah, that must be a gib victory
23:02:15bagawkI believe it is ~24$ month
23:02:21lImbusthats kinda fair
23:02:57bagawkDSL does not get that fast here
23:03:08*Bagder has an 8mbit one
23:03:09lImbushere ? where ?
23:03:23bagawk1.5mb/860kb is about the top
23:03:39quelsarukspain default dsl is 256kbps download
23:03:43bagawkCable generaly goes to 3mb here
23:03:47quelsarukmaximum is 2mbps
23:03:58quelsarukand i had 8mbps in japan
23:04:02quelsarukthis is not fair
23:04:14bagawkI am on 100mbit right at this monent though...
23:04:39quelsarukBagder: can you lend me your dsl connection?
23:04:41bagawkThanks to OSU an their OC connection :)
23:08:58bagawkBagder: I wish someone would work on tag databases :)
23:09:25Bagdersomeone does
23:09:58BagderI've written a program that builds one according to the spec in the wiki
23:10:15BagderZagor has working code that can browse such a database
23:11:01 Join Stryke` [0] (
23:11:05quelsarukdatabase in rockbox?
23:11:14quelsaruksounds good
23:11:30bagawkquelsaruk: see the wiki
23:11:45quelsaruki'm tooo lazy for that ;)
23:11:51Bagderwell, I bet a database guy could argue wheather this actually is one or not
23:11:53quelsaruki think i should read the wiki
23:12:24Bagderbut it allows you to browse songs based on their id3 tags :-)
23:14:15quelsarukwell, actually i will not use the database
23:14:48quelsaruki have my drive ordered by folders, kind of music −−> artist −−> album −−>songs
23:15:02Bagderme too
23:15:10bagawkSo do I
23:15:10quelsarukit's easy to find a song
23:15:30quelsarukbtw Bagder, do you have any experience with archos support?
23:15:33bagawkBut then with browse by tags you could search by say, year...
23:15:41Bagderquelsaruk: nope
23:16:09quelsarukmy hd began to give me problems this past week, my warranty expires in 2 days, and support hasn't answered my mails :(
23:16:44quelsarukyeah.. :/
23:17:11quelsaruki'm calling france tomorrow, just to ask for a solution
23:17:25quelsarukbut it's a expensive international call :(
23:19:25bagawkI have two 2.5" hard disks sitting around
23:19:47bagawkBoth IBM travelstarts, one 20gb, the other 15.1
23:20:27quelsarukbagawk: i have a 40gb somewhere, but i would like to have a new hd, as i paid for it and warranty is there for that
23:22:32bagawkI was thinking of putting the drives in the tivo...
23:22:55bagawkDo you know how i could put both of the drives together and make them look like one?
23:25:50bagawkDoes a RAID card work for that?
23:26:14quelsaruka raid card should work (if i understood you)
23:26:31bagawk(I know very little about RAIDs)
23:27:02quelsaruki made that with 2 SATA
23:27:22quelsarukusing the RAID built-in function of my motherboard
23:27:25quelsarukso should work
23:27:32quelsarukmayba Bagder knows better, or amiconn
23:27:36bagawkThis is going to a TIVO, no PCI slots
23:29:28quelsarukwhats a TIVO?
23:31:49bagawkThey are highly hackable :)
23:31:56bagawkThey run a linux OS
23:35:22quelsarukif they had a ethernet card...
23:35:38thegeeksoftware raid?
23:36:01thegeekshould be possible, might not be easy, but possible.
23:36:04bagawkWell, can a normal IDE chain support 3 HD's
23:36:27thegeekonly two devices per chain
23:36:48thegeekthere might be devices that does the whole shit in hardware
23:36:51thegeekbut prob damn expensive
23:36:55thegeek_not_ worth it
23:36:57 Join [IDC]Dragon [0] (
23:38:47amiconnhi again, for the nth time ;-)
23:38:51quelsarukonly 2 devices in a IDE connection
23:39:01bagawkhey [IDC]Dragon
23:39:30lImbushi Jörg
23:39:46[IDC]Dragonhi there
23:39:57[IDC]Dragonnow from cable LAN
23:41:02[IDC]Dragonhad a flaky WLAN today
23:41:23[IDC]Dragonamiconn: I have a reproduceable case here now
23:41:43amiconnFor the disappearing volume 0?
23:42:25bagawkthegeek: guess i'll just put in the single 20gb drive :)
23:42:45[IDC]DragonI had a panic, too, but not reproduceable
23:42:53amiconn[IDC]Dragon: Do you already have a more precise idea what may cause it?
23:43:12[IDC]Dragonno, but I still suspect the cacheing
23:43:24[IDC]DragonI could try some things
23:43:36[IDC]Dragonlike, splitting it into 2 caches
23:45:14amiconnThe fat driver looks rather complex to me. I'm still fighting with the FAT16 rootdir (at least I have a coarse idea how to solve that)
23:45:57[IDC]Dragonmy recipie to reproduce is: walk into the firstdir of the MMC, play something, walk to root, walk into second MMC dir, walk to root -> volume 0 gone
23:45:59amiconnI'll add a new member to the bpb, rootdiroffset. The bpb will not grow though, since I can get rid of rootdirsectors instead
23:48:11quelsaruk[IDC]Dragon: welcome to the fabulous world of the WLAN :D
23:48:12amiconnYour test case isn't completely reproducable here. Instead, I got only a part of the volume 0 root.
23:48:31amiconnWaliking into a subdir and back again fixed it
23:48:50[IDC]DragonI also had such "fixing cases"
23:49:28amiconnDid you try whether this also happens with 2x FAT32?
23:49:54amiconnMaybe this is actually a problem with the FAT16 root dir not being part of the FAT
23:50:03[IDC]Dragonno, I'm not too fond of reformatting
23:50:08amiconn(or the root dir handling in general)
23:51:17[IDC]Dragontook me quite long to get nice music onto that MMC
23:51:19bagawkTime to go
23:51:23 Join midk [0] (
23:51:31quelsarukcu bagawk
23:51:42amiconnbye bagawk
23:52:10 Quit bagawk ("CGI:IRC 0.5.4 (2004/01/29)")
23:54:00amiconn[IDC]Dragon: Really? MMCs aren't that big. Filling an 8 GB card with an USB2 card reader should take < ~10 minutes
23:54:20amiconn1 GB card of course
23:55:05[IDC]Dragonit took about 30 minutes
23:55:53[IDC]Dragonthe card reader seems not to fully exploit the interface
23:56:48amiconnThe bad thing is that my test player currently contains the 32K cluster FAT16 problem partition. Otherwise it wouldn't be a problem to use it for testing with 2x FAT32
23:57:17amiconn...comparing with 2x FAT16, and also the mixed case
23:57:47[IDC]DragonI have a reprduceable case now, that's all I need
23:57:54[IDC]Dragonbetter not mess with it
23:58:12amiconnI hope that your fixes don't collide with my fat16 changes...

Previous day | Next day