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 2002-09-04

00:00:22BoD[]because this dir filter is not the one i use most :)
00:01:14BoD[]i'd like to see a repeat one/repeat all/no repeat feature
00:01:43adi|homesend to the mail list
00:01:45adi|homeso it don't get lost
00:02:20BoD[]is there a lot of traffic on the list ?
00:02:42BoD[]because it's strange... i was subscribed and i dont receive anything anymore
00:02:54BoD[]maybe i have to subscribe again
00:04:22adi|homecheck the mail archives
00:04:27adi|homesee if you are missing something
00:05:05BoD[]well .. yes :)
00:05:16BoD[]maybe i unsubscribed by mistake
00:05:51 Join idefx [0] (
00:06:22BoD[]oh i think i was only subscribed to rockbox-sf ........ anyway
00:06:52idefxwell, i have had a file -in fact, 2, one .c, and one .h- to the apps dir, but i get undefined reference when i try to make. what more should I do ??
00:08:15 Join lohap [0] (
00:08:37lohaphi all what the hell has happen to the games plus what has happened to the wps menu
00:10:17 Join Dexter__ [0] (
00:11:02 Nick _seb_ is now known as seb-away (
00:18:34BoD[]this feature request says it all :
00:20:02lohapwell what has happened have the games been removed for good
00:20:52lohapand how the hell u change the wps
00:21:14BoD[]lohap : on the build I use, there are games :)
00:21:35BoD[]and by 'change' what do you mean exactly ?
00:21:59lohapwell from id3 to filename
00:22:12lohapbecause the newest dialy build has removed that menu
00:22:56BoD[]well sorry i dont know about this
00:23:29PsycoXuli'm not sure why the menu was removed
00:24:51lohapand why are the games removed i just like to know
00:25:41***Saving seen data "./dancer.seen"
00:27:39PsycoXulthey're still in the code
00:27:57PsycoXulthey've just been made compile-time option
00:29:02 Quit Miran ("Its not that you do not see, it is mearly that your eyes do not perceive")
00:29:10lohapstrange i think
00:29:35PsycoXulwell they take up a lot of room
00:29:44PsycoXuland there's several people who don't like them
00:29:53PsycoXulnow people have the option of having them
00:30:12PsycoXulwithout them you've got more ram free for your mp3's
00:30:28lohapwelll if u can compile source u can but i still think removing the wps menu was a dad move
00:30:42PsycoXuli really don't know why the wps menu was removed
00:30:48PsycoXuland thats unrelated to the games
00:31:18PsycoXulbut the WPS is now just fully custom
00:31:44lohapyea but u need to edit the conf and compile which i cant do
00:31:54PsycoXulyou don't need to compile
00:32:06lohaphow do u do it then
00:32:11PsycoXulyou just make the default.wps according to the doc and put it in your /.rockbox/ dir on your archos
00:32:27PsycoXuland then you can make whatever other .wps's you like and also put them on there and select them to use them
00:32:27 Quit idefx ("BitchX-1.0c19 -- just do it.")
00:33:15PsycoXul(in the browser, with mp3/m3u filter turned off)
00:36:13BoD[]anyway i have to go to bed !
00:36:18BoD[]see youm byeeee
00:36:56 Quit BoD[] ("MBLELOPPPPPPPPPPPPP POWERRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrritu)
00:36:59PsycoXulBjörn removed the WPS display setting the day before yesterday
00:37:09PsycoXulfrom the menu
00:37:17TotMacherwhat about an excel file viewr ?
00:37:24TotMacheror power point files
00:37:34PsycoXulTotMacher: code it :p
00:39:29lohapwhich dir must eh default.wps located in the root dir
00:39:30TotMacherbut, i also think, that 400 files are too less
00:39:46TotMacher1000 would be nice
00:49:09 Join Miran [0] (
00:49:31Miranhey, does anyone know when rockbox is going to be updated so that it recovers from read errors?
00:50:04TotMacherdoes anyone here use the remote control ?
00:50:44 Part Dexter__
00:54:01 Quit manel ("Leaving")
00:56:56 Join eye69 [0] (
00:57:02eye69Hello people.
00:58:44TotMachera spectrum analyzer would be very kewl !
01:06:04 Quit lohap ()
01:09:07 Quit Miran ("Its not that you do not see, it is mearly that your eyes do not perceive")
01:29:25 Quit TotMacher ()
02:17:59 Join ddb [0] (
02:19:21ddbHi folks, I have a question about the newest version of the WPS config file...
02:23:25 Part ddb
02:25:46***Saving seen data "./dancer.seen"
02:35:39 Quit pimlottc (Read error: 110 (Connection timed out))
03:09:43 Join ddb [0] (
03:10:04ddbquestion about WPS...
03:14:28 Part ddb
03:24:26 Quit datazone (Read error: 110 (Connection timed out))
03:37:28 Join elinenbe [0] (
03:39:38 Join Tigren [0] (
03:46:37 Join merwin [0] (
03:46:58Tigrenhow you doin
03:51:38merwinpretty good
03:52:49 Join mecraw12 [0] (~mecraw@
03:53:40merwinhey mecraw
03:56:07mecrawlooks like you've been busy
03:56:27merwinyup, a 5 hour flight will do that to you
04:00:32merwinhow long does a charge take?
04:01:03mecrawi can't tell, rockbox says mine is at 100% after an hour
04:01:23mecrawbut i think it takes about 6 hours to be completely charged
04:03:02merwinmine always says HD error sometime after 5 or 6 :)
04:03:57mecrawthat doesn't sound like a good charge
04:04:19merwinapparently it's a problem with the players
04:04:36merwinI just bought 1700mAh batteries and a new charger, so maybe it'll go away
04:05:01 Quit Tigren (Read error: 54 (Connection reset by peer))
04:08:09 Join Tigren [0] (
04:11:37 Quit merwin ()
04:13:06 Join datazone [0] ([eMuNJtk1X@
04:24:19 Join merwin [0] (
04:25:47***Saving seen data "./dancer.seen"
04:27:53 Join pimlottc [0] (
04:40:11 Quit Synthe (Read error: 54 (Connection reset by peer))
04:40:12 Join Synthe` [0] (
04:40:22 Nick Synthe` is now known as Synthe (
04:46:54 Nick seb-away is now known as seb-sleep (
04:48:23 Join ddb [0] (
04:48:50ddbhi merwin
04:50:29 Part ddb
04:54:58 Quit edx|sleep (Read error: 110 (Connection timed out))
05:14:49 Join g003y [0] (
05:16:50g003ywas adimas around todya?
05:27:26 Part g003y
05:27:49 Part Tigren ("Client Exiting")
05:32:10 Nick elinenbe is now known as elinenbe|sleep (
05:32:51 Nick merwin is now known as merwin|zzzz (
05:39:03mecrawmerwin|zzzz: early to bed, early to rise?
06:09:31adi|homemakes a man get herpes
06:22:15merwin|zzzzmecraw: nah, not feeling well
06:22:20merwin|zzzzmecraw: dizzy as hell :P
06:22:41adi|homereal men love 48 hr shifts
06:22:45adi|homesuck it up little girl
06:22:54adi|homehere... i make you smile
06:25:51***Saving seen data "./dancer.seen"
06:26:43adi|homesee.. now you feel more awake
06:27:29merwin|zzzzno, i'm perfectly awake, just dizzy as hell
06:30:05merwin|zzzzwhat's your obsession with 80 char line? :)
06:32:00adi|home1. its how the rest of the code is (98% anyways)
06:32:11adi|home2. lines > 80 chars are just signs of laziness
06:32:12 Quit Synthe (Read error: 54 (Connection reset by peer))
06:32:17 Join Synthe [0] (
06:32:25adi|home3. i follow the example of the code police ;)
06:33:42 Join edx|sleep [0] (
06:45:47merwin|zzzzadi|home: well, over 80 chars aren't necessarily a sign of laziness.. when you use 4 space indentions, you can get some really short lines :)
06:47:16*merwin|zzzz got some new 1700 mAh batteries
06:47:25merwin|zzzzenergizer, so they're decent :)
06:51:42merwin|zzzzmy archos made a very strange noise just now
06:51:52merwin|zzzztime for a chkdsk
06:52:01 Nick merwin|zzzz is now known as merwin (
06:54:54 Quit Synthe (Read error: 54 (Connection reset by peer))
06:54:54 Join Synthe` [0] (
06:59:59 Nick Synthe` is now known as Synthe (
07:04:16 Join LinusN [0] (
07:06:35mecrawhi Linus
07:07:12merwinhey linus
07:09:28merwinso is charging with 9v ok? I've got a charger with voltage setting on it from 4v to 12v and am wondering the highest voltage... I read the battery faq, and it seems that 9v is a good max
07:09:55LinusNelinenbe|sleep: raised the default contrast to a very dark level on my recorder :-(
07:10:29LinusNthe archos travel charger is 12V
07:11:00mecrawi set my contrast at 40, 32 was way too dim
07:12:19LinusNhehe, the units ave very different in this aspect
07:12:48LinusNit doesn't matter that much. The most important thing is that the defaut is readable on all units
07:27:00LinusNwhy did you remove the last_button != BUTTON_LEFT check?
07:27:13LinusNit was there for a reason
07:27:50merwinLinusN: really? it screwed up track changing on the player
07:28:01LinusNon the player?
07:28:02merwinLinusN: it was basically non-existant :)
07:28:13merwinLinusN: yeah, you had to press NEXT NEXT to get it to change tracks
07:28:19merwinand really quickly mind you
07:28:41LinusNok, but maybe you should have done it only on the player?
07:28:57merwinI didn't know it was enabled for a reason... why would it be there?
07:29:22LinusNyou can leave the menu with the LEFT key on the recorder
07:29:42LinusNand the WPS gets the key-up event and thinks that the user pressed left
07:29:56merwinahh, i'll throw it in for the recorder then :P
07:30:10LinusNwe really need a better handling of this stuff... :-(
07:30:35merwinthe keypad you mean?
07:31:22LinusNyeah, all these key combinations and key-release events
07:32:43*merwin is going to have an after-commit cigarette :P
07:33:01*mecraw holds his breath
07:33:04*merwin also wonders why lastbutton doesn't work right on the player
07:33:23merwinmecraw: what does mecraw mean?
07:34:01merwinyou know, on a 15GB of data via USB, a chkdsk sure takes a damn long time
07:34:21 Nick merwin is now known as merwin| (
07:34:27 Nick merwin| is now known as merwin|puff (
07:36:01mecrawmerwin|puff: it's a misspelling of meecrob
07:38:28merwin|puffthe thai food?
07:38:32 Nick merwin|puff is now known as merwin (
07:38:50merwinI'm amused by my new muffler... it sets off car alarms in the parking garage at my work :-)
07:38:55mecrawyeah, i took it from the sh*t episode of South Park
07:43:58LinusNrolo really speeds up my development cycle, since i don't have to restart when i have replaced ajbrec.ajz
07:44:32mecrawit's a godsend
07:44:45 Nick merwin is now known as merwin|dizzy (
07:44:59mecrawmerwin|dizzy: stop spinning, then
07:45:52mecrawmy only problem with all the new .wps and .ajz is having to turn off filtering
07:46:12merwin|dizzymecraw: wish i could stop being dizzy :P can't sleep when i feel like fainting (sounds weird)
07:46:42mecrawshowing hidden files for .rockbox, makes all of the windows crap directories visible
07:47:23merwin|dizzyLinusN: what do you think about a setting for filtering that has rockbox supported files only
07:47:40LinusNyeah, we would probably need that
07:48:58mecrawthat, or just clean out all of the crap... that'll be my choice
07:49:23mecrawhow about a question mark icon for files we don't handle?
07:49:53merwin|dizzymecraw: we had one, but it clutters one up more than it helps
07:50:30merwin|dizzymecraw: it's actually easier to not have any icon for non-supported files (at least on the player)
07:51:03mecrawon the recorder it looks like a nested folder
07:51:23merwin|dizzywhatever happens, we need to NOT enable it on the player :-)
07:53:19merwin|dizzyok, time to try sleeping again
07:53:34 Nick merwin|dizzy is now known as merwin|zZzZz (
07:53:42mecrawme too, good night
07:59:05 Join rwood [0] (
08:00:13rwoodLinusN: were you able to run the i2c software?
08:00:38LinusNi haven't had the time yet
08:01:08rwoodok - i'll try to get the player log by the end of the week
08:01:20 Join Bagder [0] (
08:01:25LinusNyo Bagder!
08:01:49Bagder...67 unread mails
08:03:01Bagder32 voters so far
08:04:30Bagderthose html mails screw up my vote count script
08:04:40LinusNi hate html mails
08:05:33LinusNi hate when people think they know best what font and font size *I* should have when i read their mails
08:05:58LinusNand what color it should have, and blablabla...
08:06:20LinusNand when 10 words of text use up 1Kbyte
08:07:02LinusNhehe, the worst habit i have seen so far is on a Matrox beta product list i am on
08:07:11LinusNall people send html mails
08:07:26LinusNand for every bug report we send screen shots as attachments
08:07:43LinusNand for every reply, people *include* the attachment!!!!!!
08:07:57LinusNevery mail is >100Kbytes
08:10:55merwin|zZzZzmy archos has bad clusters....
08:11:04merwin|zZzZzwhich i assume means physical error
08:11:09rwoodmerwin|zZzZz talks in his sleep
08:11:30 Nick merwin|zZzZz is now known as merwin|grr (
08:11:40merwin|grrso anyone here bought their unit at best buy?
08:12:01LinusNis best buy a store?
08:12:07merwin|grrLinusN: yeah...
08:12:20rwoodmerwin|grr: can you format?
08:12:35merwin|grrrwood: yeah, but i'd rather trade it in for a brand new unit at best buy if it's covered under the warranty :)
08:12:58rwoodmerwin|grr: better now than later!
08:13:10LinusNbe sure to brush away the marks from when you dropped it on the floor :-)
08:13:29rwoodand delete rockbox
08:13:29merwin|grractually, suprisingly enough, i haven't dropped it :)
08:13:46LinusNyou should definitely remove rockbox from it
08:13:48merwin|grrrwood: rockbox doesn't void warranty
08:13:58LinusNno it doesn't
08:14:20LinusNbut you will spare yourself the trouble of explaining that to the Best Buy people
08:14:29merwin|grrLinusN: hehe
08:15:10merwin|grrhopefully sometime in the next 2.5 years archos will stop selling the players, and something will "accidently" happen to it and i'll get it replaced with a recorder
08:15:11LinusN"So your PC is broken. Did you run Linux on it?"
08:15:21LinusN"Sorry, only Windows is allowed"
08:15:32merwin|grrLinusN: heh, yeah
08:15:34LinusN"Linux isn't covered by the warranty"
08:16:38merwin|grrso far it's found bad clusters in 15 files
08:16:59merwin|grrwoo, there's a best buy not 15 miles from where i am in chicago right now
08:17:10merwin|grri'd better remember to take out my nice new batteries too
08:17:27LinusNmerwin|grr: battery status?
08:17:38merwin|grrLinusN: scuse me?
08:17:52LinusNi wonder how your batteries are
08:18:12merwin|grrLinusN: oh, brand new energizer 1700 mAh's...
08:18:17merwin|grrjust got 'em today
08:18:22merwin|grrprobably just finished charging too
08:20:45LinusNbe back in a f....? fecond?
08:20:59LinusNhe has a lifp?
08:22:55LinusNah, thx
08:23:22LinusNonce upon a time i was good at english
08:23:32LinusNthose were the days
08:23:45 Join PH [0] (
08:23:57 Quit PH (Client Quit)
08:23:57Bagderyou mean when you were still young and agile? ;-)
08:24:03 Join PH [0] (
08:24:09LinusNi was?
08:24:14LinusNi can't remember
08:24:36rwoodas oposed to old and fragile?
08:24:44 Nick PH is now known as PHsoft (
08:25:32Bagderhehe, correcto
08:25:36LinusNrwood: yes, old, fragile, bad memory
08:25:41Bagder32 people counted
08:25:46Bagder 1: design 6 (28 points)
08:25:46Bagder 2: design 24 (22 points)
08:25:46Bagder 3: design 7 (18 points)
08:25:49LinusNway over 30 years old
08:25:53***Saving seen data "./dancer.seen"
08:26:08rwoodclose to double
08:26:10LinusNBagder: you are falling behind
08:26:29LinusNpity, i like your design
08:26:45Bagderwe could do a "Björn" ;-P
08:26:58merwin|grrBagder: hahahaha
08:27:14LinusNi don't want the names on the shirt
08:27:25Bagderwell, plenty of people do
08:27:52LinusNthe list has grown since then
08:27:56merwin|grrLinusN: too many people get added to the list :)
08:28:08Bagderthere's probable ten more names now
08:28:16merwin|grrLinusN: names on the shirt are irrelevant until the project is 100% completed, which it will never be
08:28:37Bagderwe could ship a roll of tape and a pen with them!
08:28:45Bagder"expand it yourself" ;-)
08:29:20LinusNprinting such a large design is costly
08:29:34merwin|grrdoes a "bad cluster" in NT/Win2k mean hard drive failure or just cross-linked file or something
08:29:44merwin|grrLinusN: yes
08:29:49Bagderor even, when we squeeze the lsit into a suitable size, no one will be able to read the names on the shirts B-)
08:30:02merwin|grrLinusN: i found a website a bit ago that lets you design it interactively and it tells you how much it'll cost based on the size and color types
08:30:40PHsoftHey I just downloaded the new rockbox and now I can't find the 'while playing'option anymore
08:30:50merwin|grrPHsoft: you sure?
08:30:51LinusNPHsoft: yes, it's gone
08:30:58merwin|grryeah, i forgot
08:31:01merwin|grrcustom wps
08:31:10*merwin|grr is extremely stupid
08:31:19merwin|grrPHsoft: loadable custom wps files
08:31:22Bagderwe've gone mad
08:31:54merwin|grrBagder: i don't know if we should trust your count on the t-shirts... since yours IS in the top 3 :-)
08:32:09merwin|grrBagder: which one of you and Zagor are older?
08:32:18LinusNBagder is *way* older
08:32:26Bagdercan't you tell by my great wisdom? ;-P
08:32:28LinusNhe is very old in fact
08:32:39merwin|grrBagder: age? Zagor looks about 25
08:32:51BagderI'm 31, Zagor's 29
08:33:01LinusNand i'm 33
08:33:15LinusNi think
08:33:18merwin|grrBagder: Zagor ages well :-) (haven't seen a pic of you)
08:33:20LinusNcan't remember
08:33:23*merwin|grr is a young nubile 22
08:33:28 Nick merwin|grr is now known as merwin (
08:33:39rwood58 here - should change nick to greybeard
08:33:40LinusNmerwin|grr: how do you know that zagors picture is recent?
08:33:58LinusNhe has a beard nowadays
08:34:07Bagderrwood: ;-)
08:34:26merwinLinusN: well, it was from the days of Archos stacking
08:34:28 Quit PHsoft ()
08:34:29LinusNa grey beard
08:34:38LinusNlong, grey beard
08:34:51Bagderyeah, LinusN and Zagor can be seen on the devcon pictures
08:34:58BagderI managed to avoid getting caught
08:35:03LinusNam i on those pictures?
08:35:22merwinLinusN: yeah... you look very sweedish :-)
08:35:30merwinerr, s/ee/e
08:35:48Bagderbork bork bork ;-)
08:35:59LinusN"make a looong face to be sweedish"
08:36:20merwinVery much like the Ericsson people I know :-) They're very stereotypical... funny a bit
08:36:20LinusNyeah, i can see my nose there on one picture
08:37:58LinusNi worked at Ericsson for 6 years :-)
08:38:00merwinso zagor has a beard? i can't see that as looking very good
08:38:05*merwin looks around for the code police
08:38:10merwinLinusN: really? heh, i'm sorry
08:38:43*merwin isn't quite fond of Ericsson. They do throw good parties though
08:38:54merwinHejsan alla barn - what's that mean?
08:39:02LinusNhello, all kids
08:39:30LinusNmy "home page"?
08:39:44LinusNi haven't got around to do that yet...
08:39:52merwinLinusN: yeah
08:39:58LinusNand, no, zagor doesn't have a beard :-)
08:40:24merwindamn travel more than me :P
08:40:58Bagderhehe, but that with your frequency
08:41:07Bagderbut not
08:41:25*merwin wants to get to Europe again sometime...
08:41:34merwinall i got to see was UK and Czech republic
08:45:20merwinwoo, kama sutra on showtime :P
08:53:32Bagderhey hes
08:55:06HesI wouldn't like my name on a shirt I'd wear 8-)
08:59:14adi|homebtw.. whats the tshrit vote count at?
08:59:23adi|homenm.. scrolled back...
08:59:55rwoodwhat is the voltage on fully charged batteries?
08:59:59LinusNi give up!
09:00:05LinusNi give up!
09:00:06DBUGEnqueued KICK LinusN
09:00:06LinusNi give up!
09:00:06LinusNi give up!
09:00:07***Alert Mode level 1
09:00:32LinusN"You know what would be cool.... "
09:00:32adi|homejust and idea
09:00:42LinusN"A seperate forum for Rockbox questions,
09:00:42LinusNproblems, concerns."
09:00:55LinusNWE HAVE THAT!!!!!!
09:01:06adi|homei think on the tshirst.. we should be able to vote for a front from one shirt, and a back from antoher
09:01:07adi|homeas part of one vote
09:01:07adi|home.. :)
09:01:07adi|homeick.. electrical storm..
09:01:07adi|homeim bound to get bounced in 10 mins
09:01:12CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
09:01:12*Bagder rolls around on the floor laughing
09:01:28adi|homedude.. tell me you have a link to that?
09:01:44Bagderadi|home: I don't think separate back and front vote would be very cool
09:01:53Bagderit would only get us totally different ones
09:02:17Bagderthat wouldn't look good together
09:03:05adi|homei like the idea of the rockbox logo (like my one submission) on the chest from
09:03:10adi|homefront rather
09:03:14adi|homeand a solid design on the back
09:03:51LinusNBagder: are games and screensavers not included in the daily builds?
09:04:02adi|homeokay.. can i respond with my normal level of sarcasm to that damn post?
09:04:03Bagderthey're not?
09:04:08adi|homeno they should be
09:04:24adi|homeassuming the new config script was copied over
09:04:37LinusN"there are features missing from the new builds, it's not your player. They took out screensavers/games to save room. "
09:05:28adi|homedamnit... that's not right..
09:05:28adi|homeill dl and see.. give me a min
09:05:50Bagderthey're not included
09:05:58Bagderthe configure script is not done correct
09:06:09BagderI think we should default to all-included
09:06:13adi|homedefine 'is not done correct'
09:06:16Bagderand disable specific things
09:06:17LinusNdefault should be yes, imho
09:06:39Bagderadi|home: 'configure update' should work and do the Right Thing ;-)
09:06:50adi|homeit is yes by default on the config.
09:07:04BagderI think you should reverse the define
09:07:08Bagderset the define to disable
09:07:21adi|homei understand what you mean, but thats different from what you said ;)
09:07:37Bagderin your view, yes ;-)
09:07:46LinusNBagder: the automated build does "configure update"?
09:08:05BagderI need to check
09:08:29Bagderhm, no
09:08:36Bagderthey don't
09:08:41LinusNi just ran "configure", and it defaults to yes
09:08:52adi|homethank you
09:08:54Bagderwhen you did configure update?
09:08:57adi|hometold ya it did
09:09:01LinusNthe automated builds should run from scratch, don't they?
09:09:11Bagdernot the configure part
09:09:19LinusNi never run "update"
09:09:25LinusNi don't see the point
09:10:07adi|homeit all depends on how the configure script gets run..
09:10:08***Alert Mode OFF
09:10:14adi|homebut i understand what bagder means...
09:10:21adi|homeits an easy fix
09:10:27adi|homebut i don't know if i agree with it to be honest
09:10:51LinusNdo we have an explanation to why the games/screensavers are excluded?
09:11:01adi|home#define DONT_USE_GAMES just looks funny to me
09:11:12adi|homeit was requested at some point
09:11:12adi|homeand it was easy
09:11:16 Part Bagder
09:11:18adi|homeso i did it to shut whoever it was up :)
09:11:37LinusNno, i mean in the daily builds
09:11:40adi|homeright now it doesn't save us much on space.. ~4secs of buffer
09:11:48adi|homeoh.. that .. no
09:12:03adi|homebut i assume its because the configure script is treating it as a no to including them...
09:12:12adi|homewhich means that it's not really being run as if from the command line
09:12:18adi|homeelse it would default to being included
09:14:12merwinPosts: 1 Posted - 4 September 2002 1:15
09:14:12DBUGEnqueued KICK merwin
09:14:12merwinDamnit people.. god forbid you go and read the FAQ about the project! I mean... Its not like its Frequently Asked Questions or anything.
09:14:13***Alert Mode level 1
09:14:17merwinnice adi :-)
09:16:07adi|homethank you
09:16:18adi|homemy 'dealing with stupid ppl's bullshit' buffer is overflowing
09:16:30merwinadi: *grin*
09:17:21merwinSo i don't think i ever got an answer to this (if anyone knows). When chkdsk in NT/Win2k says "Bad Cluster" do they mean hard disk failure or cross-linked file, etc
09:18:26adi|homeno idea
09:19:00adi|homethere linus
09:19:06LinusNi think that the HD reported a read error
09:19:13LinusNadi|home: found it
09:19:32merwinLinusN: cool, off to best buy in 7 hours then :-) Cash in that warranty!
09:19:48LinusNbetter do that
09:20:15merwinBasically, almost every single one of my 400 or so SNES ROMs have bad clusters
09:20:19 Join Bagder [0] (
09:20:23merwinhey badger
09:20:31Bagderweirdness happens
09:20:42 Join matsl [0] (
09:20:49Bagderso where were we?
09:21:00merwinso when is the archos divx recorder going to be available?
09:21:03BagderI think no extra defines should mean default-build
09:22:07adi|homebagder so in other words you want a
09:22:40adi|home#define DEFINE_ME_TO_TURN_OFF_GAMES <code> #endif
09:23:03BagderDISABLE_GAMES perhaps
09:23:06adi|homei dunno.. just feels.. odd to me...
09:23:30adi|homei tend to look at defines as "include everything following"
09:23:30BagderI don't like to force people to re-run configure to maintain the same build
09:23:49LinusNBagder: the auto scripts do run "configure update"
09:23:57adi|homebut i think the problem is with how the config is running.. not the defines...
09:24:07adi|homei dunno...
09:24:14***Alert Mode OFF
09:24:18adi|homeand here comes the storm...
09:24:19adi|homeill be gone in a little bit folks
09:24:37LinusNand configure update sets the options to default
09:24:53merwinadi|home: hmm?
09:24:55adi|homewhich should mean 'turn my games on"
09:25:07adi|homemerwin: electrical storm moving in
09:25:35LinusNmaybe the auto scripts don't update the configure script?
09:25:47merwinadi|home: ahhh... not good
09:25:49Bagderthey don't
09:25:59Bagderbut I still think the configure has wrong approach
09:26:07 Nick merwin is now known as merwin|zZzZ (
09:26:07DBUGEnqueued KICK merwin|zZzZ
09:26:09Bagdereven if the cron job is wrong too
09:26:09 Nick merwin|zZzZ is now known as merwin|zZzZz (
09:26:09DBUGEnqueued KICK merwin|zZzZz
09:26:13LinusNyou mean disable vs enable?
09:26:39LinusNexclude=no rather than include=yes
09:26:41Bagderwithout me touch my configure, I now all of a sudden have disabled all my games and screensavers
09:26:52Bagdernot quite
09:26:53adi|homejust seems to me that if we treat these defines as you imply they should be.. its diff from all other instances of our defines
09:27:03Bagderno flags = default build
09:27:04LinusNi agree
09:27:07Bagderflags = changed build
09:27:13mecrawWhy isn't Rockbox listed on
09:27:16LinusNBagder: perhaps
09:27:16Bagderwhatever the default is
09:27:19adi|homeall our current cases are of the #ifdef LCD_BITMAP <code> #else <code> #endif kind
09:27:38PsycoXulthere's no #ifndef's? you sure?
09:27:44merwin|zZzZzadi|home: there are some ifndefs
09:27:49Bagderadi|home: those are hardware feature defines
09:27:57Bagdernot feature-options
09:28:18LinusNconfig.h could easily translate the defines for us
09:28:51LinusN#ifdef DISABLE_GAMES
09:28:58LinusN#undef USE_GAMES
09:29:08LinusNor something
09:29:33adi|homehmmm.. that would seem to fit more in line with the logic we have over all.. but is a bit circuitous..
09:29:42adi|homewhat do you think Bagder?
09:29:42Bagderwell yes, if USE_GAMES is the preferred method which I can't see why it is
09:30:06LinusNBagder: i like positive logic more than negative
09:30:12BagderI think we mix hardware-features with compile-time feature-options
09:30:37LinusNi like "include" better than "exclude"
09:30:41Bagderfine with me
09:30:52BagderI'm more concerned for the compiler flags and makefiles
09:30:52adi|homewell.. we have "#ifdef LCD_PROPFONTS" is that hardware or feature-options?
09:31:01LinusNi like HAVE better than LACKS
09:31:08Bagderadi|home: they modify the default
09:31:10Bagderthen they are set
09:31:16Bagderdefault is default
09:32:03BagderLinusN: this isn't HAVE or LACKS, this is SHUTOFF or DEFAULT
09:32:25LinusNi think you get my point
09:32:30BagderI do
09:32:33Bagderand I say I'm fine with it
09:32:43LinusNin *the code*, i want USE_xxx instead of DONT_USE_xxx
09:32:46BagderI'm not fine with requiring compiler options for default actions though
09:32:53LinusNBagder: i agree on that
09:33:12merwin|zZzZzheh... Neo Jukebox without hard drive costs $199, Neo Jukebox with 10GB hard drive costs $209.... hmmm....
09:33:15LinusNtherefore i suggest that config.h takes care of that
09:34:11Bagderno objections from me
09:34:19LinusNwhats a neo jukebox?
09:34:53Bagderthe daily build will now do a configure update properly
09:34:53 Join adiamas [0] (
09:34:57merwin|zZzZzLinusN: like the archos jukebox, different company7
09:35:12LinusNBagder: i fix the config.h
09:35:28adiamasback.. what i miss?
09:36:46LinusNBagder: config.h was not a good idea...
09:37:07LinusNmaybe an appconfig.h?
09:37:19LinusNconfig.h is in firmware/
09:37:31adiamasummm.. what am i missing?
09:37:52LinusNadiamas: we are talking about the USE_ vs DISABLE_
09:38:03merwin|zZzZzi go for use
09:38:03adiamasahh.. okay...
09:38:21adiamasi vote for USE, but i understand Bagder's side
09:38:25LinusNshould i create appconfig.h?
09:38:32adiamasso im happy either way
09:38:37Bagderperhaps feature.h ?
09:38:45adiamasnot if its just for this
09:38:51adiamasif we need stuff later, then yeah.
09:39:13adiamashad a brain storm when thinking about the wps
09:39:28adiamashow does the idea of a configurable filter sound?
09:39:37LinusNin the wps????
09:39:39adiamasusers could set levels/values as they see fit
09:39:42adiamasno no no
09:39:55adiamasseperate config
09:40:28adiamas[Mp3 Only]
09:40:50LinusNsound filter?
09:41:15adiamas[PlayList and Mp3]
09:41:24adiamas.mp3 .m3u
09:41:51adiamaswe could have it basically be a list of things that are to _not_ be filter and things that _are_ to be filtered
09:41:59adiamasie: * is all
09:42:07adiamas!.mp3 is everything that is Not an mp3
09:42:11LinusNthat list will easily grow when we add file types
09:42:19CtcpIgnored 3 channel CTCP requests in 11 minutes and 57 seconds at the last flood
09:42:19*adiamas is brainstorming
09:42:29adiamasand we can have defaults
09:42:35adiamasand the users can add/remove at will
09:42:38adiamasor just leave it be
09:43:04adiamastheir choices over ride our defaults
09:44:27LinusNyou mean in .rockbox/config?
09:46:07adiamasthis case it would be
09:46:51*adiamas wonders if he is making sense
09:47:56miahi was running 1.3 today... it worked great. one thing i really think needs added is a way to lock the keys
09:48:07miahit kept skipping to the next song when i had it in my pocket
09:48:19Bagderso use the key-lock ;-)
09:48:29Bagderr... t... f... m! ;-O
09:48:59 Quit adi|home (Killed (NickServ (Nickname Enforcement)))
09:49:19 Nick adiamas is now known as adi|home (
09:49:34Mode"#rockbox +o adi|home " by ChanServ (ChanServ@services.)
09:49:42Topic"Make your voice heard!" by adi|home (
09:49:49Mode"#rockbox -o adi|home " by adi|home (
09:50:28adi|homewe really need a 'All you ever wanted to know about Rockbox but we made fun of you for asking' doc
09:50:54adi|homethats what we need
09:51:08Mode"#rockbox +o adi|home " by ChanServ (ChanServ@services.)
09:51:29Topic"Make your voice heard! (The gods demand an AYEWTKARBWMFOYFA doc)" by adi|home (
09:52:12Mode"#rockbox -o adi|home " by adi|home (
09:54:18LinusNi was thinking...
09:54:33LinusNmaybe we should add something in the FAQ like this:
09:55:23LinusNWhen will you fixed the bug i reported to <insert your favourite web forum>?
09:56:03LinusNanswer: since you haven't told us about the bug, we will vener fix it.
09:56:11adi|homenormally i would say sure...
09:56:22adi|homebut if they are posting to those forums...
09:56:22adi|homethey don't read the faq
09:56:31adi|homestupid is as stupid does
09:57:28adi|homeand im getting tired of holding hands for ignorant users
09:57:39adi|homeppl that are new, cool.. im more then happy...
09:57:45adi|homeppl that want to learn.. wonderful...
09:57:55adi|homeppl that want to call us fuckwits and losers...
09:58:09adi|homei send them to Bagders house
09:58:09adi|homebut their cool...
09:58:17adi|homeignorant bastards.. well... fuck off
09:59:59adi|homei think every project has one person.. one really odd person.. that has no patience for fools, and treats them as they deserve to be treated... and though the outside world despises that one person.. the group body cherish's them for saying the things they wish to say but cant for the good of the project
10:00:10adi|homei humbly volunteer to be that asshole...
10:00:15adi|homeI mean.. person
10:00:44Bagder"oh, you're a newbie and want to ask us a question, sure, go talk to adi" ;-)
10:00:45adi|homeunless anyone wants to help
10:00:57adi|homei could use a subordinate
10:00:58*adi|home smirks
10:01:03adi|homelol yup.. send em to me..
10:01:30*adi|home smacks them with a LART, shows the FAQ url down their throat and sends them on their way
10:01:37*adi|home yells out 'Next!'
10:01:39*adi|home smacks them with a LART, shows the FAQ url down their throat and sends them on their way
10:01:40*adi|home yells out 'Next!'
10:01:42*adi|home smacks them with a LART, shows the FAQ url down their throat and sends them on their way
10:01:44*adi|home yells out 'Next!'
10:02:50*adi|home scrolls back and figures out that he owns 60% of the irc logs.. but only 3% of that 60% is actually useful info/advice/sarcasm
10:03:49 Quit rwood ()
10:04:12*adi|home crowns himself "King of the dually LART'd"
10:04:54*adi|home names Bagder "Lord High Mucky Muck of users not worthy of LART-erdom"
10:05:22*adi|home taps LinusN on the shoulder and asks what position he wants.....
10:05:40 Nick Bagder is now known as LordMuckyMuck (
10:05:40adi|homeoh oh oh...
10:06:14*adi|home dubs LinusN 'Knight of Cunning Linguists and Borks'
10:06:24adi|homeget it.. cunning linguists!
10:06:32*adi|home laughs his ass off...
10:06:50LinusNKnight of the Cunnilinguists?
10:06:50 Nick adi|home is now known as KingOfDuallyLart (
10:06:57KingOfDuallyLartno no no
10:07:03KingOfDuallyLartCunning _linguists_
10:07:04LinusNi'm all for that!
10:07:20KingOfDuallyLartyou use your mouth purtier then a 20$ whore
10:07:56 Nick LinusN is now known as SirKnowalot (
10:07:56DBUGEnqueued KICK SirKnowalot
10:08:15KingOfDuallyLartbehold the royal court
10:09:47KingOfDuallyLartnow.. when Zagor shows up.. we'll get him a title too
10:10:24 Nick KingOfDuallyLart is now known as KingOfTheLarted (
10:11:25 Part SirKnowalot
10:15:09 Join bobTHC [0] (
10:15:16bobTHCgood morning
10:19:10CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
10:19:10*LordMuckyMuck grins
10:19:15 Nick LordMuckyMuck is now known as Bagder (
10:19:33Bagdermatsl: did you intend to mail us blank votes?
10:19:46 Nick KingOfTheLarted is now known as adi|home (
10:20:21 Join LinusN [0] (
10:20:33LinusN*boooooom* usb-storage panic
10:21:00adi|homeick.. what happened?
10:25:32LinusNthe usb-storage got an unexpected interrupt and froze the whole box
10:25:48adi|homeis wrong
10:25:54***Saving seen data "./dancer.seen"
10:26:09adi|homeits no longer Sound Settings
10:26:13adi|homewe have wps info there..
10:26:15adi|homedeep discharge
10:26:20adi|homeneeds a new title is all
10:26:25LinusNgo update
10:26:38Bagdercheck out the player manual for kicks
10:26:45Bagderit hasn't been touched since the 1.2 release
10:26:53adi|homei am not updating the manual anymore
10:26:56adi|homeeverytime i did
10:27:00adi|homethe changes didn't take
10:27:07Bagderthey did
10:27:11adi|homecause we had different branches for the diff releases
10:27:28Bagderno, we copied the manual to manual-1.2
10:27:31Bagderfor the 1.2 release
10:27:39adi|homei fix then
10:29:11adi|homewe seriously need a new screenshot for that page too
10:29:28LinusNwe probably need a text-file manual to accompany the releases
10:29:50Bagderwe can't even make a web one, how can we make a text one? ;-)
10:29:52adi|homei volunter merwin
10:30:17LinusNBagder: the web one is more work to update
10:30:27LinusNscreen shots, links...
10:30:38Bagderwell yes, if you want that
10:30:45Bagderbut people don't update the plain info either
10:30:50Bagderthe easy stuff
10:30:51eye69Anyone have the latest CVS for the recorder before the fonts went back to "normal"?
10:30:55Bagderso I don't buy that argument
10:31:01eye69I couldn't find an archive on the page
10:31:10Bagdereye69: you mean propfonts enabled?
10:31:18LinusNwhat if i made a screendump feature on the target?
10:31:20eye69Bagder: Yeah. I really like those.
10:31:34eye69Bagder: The screen fits so much more.
10:31:37Bagdereye69: they're still possible to enable if you build your own
10:31:45BagderI like the propfonts too
10:32:01Bagderwe're aiming on introducing loadable fonts soonish as a replacement
10:32:01eye69Bagder: Ok...I'll have to build the compiler then...
10:32:12eye69Okie, sounds nice.
10:33:47adi|homeokay.. bed time
10:33:51adi|homesee you all in a few hrs
10:33:59Bagdernight adi
10:34:08eye69Bagder: I got a CVS from Zagor a while back, with them enabled. It was before seek though.
10:34:31Bagderright, anyone with the dev environment setup can build one
10:34:45*Bagder runs to get more coffee
10:35:24adi|homeeye69: recorder or player you need?
10:36:51adi|homeeye69: ?
10:37:02adi|homegoing once
10:37:02adi|homegoing twice
10:37:13adi|homeand adi is sold to his bed for happy dreams
10:37:15LinusNrecorder, of course
10:37:26LinusNpropfonts, remember?
10:37:40adi|homei don't use them
10:37:44adi|homehavne't played with thim.. so no idea
10:37:53LinusNeye69 wanted propfonts
10:38:04adi|homebuilding for him now
10:38:21LinusNyou can be a nice person, when you try really hard :-)
10:38:34adi|homeim generally a nice person
10:38:50adi|homei just think we are too far removed from the survival of the fitest
10:39:02adi|homei think they should introduce bears and tigers into grade schools
10:39:10adi|homethat whould thin the herd nice and quick
10:39:51adi|home when he comes back
10:40:50eye69adi|home: recorder
10:41:03LinusNwhy make bears and tigers go to school? i don't get it. :-)
10:41:06eye69adi|home: Ahhh, thank you very much.
10:41:28adi|homeyour welcome
10:41:36adi|homeand tigers and bears need to eat too LinusN :)
10:41:38adi|homenite all
10:44:20BagderI wish xchat had dabbrev-expand
10:51:58eye69Ahhhh, propfonts...nice
10:52:27Bagderyes, me like em too
10:52:37matslBagder: Time to try chatting through Emacs then ;-)
10:52:42eye69That ROLO-feature is cool
10:52:53Bagdermatsl: yes indeed
10:53:30LinusNmatsl: blank voting is so lame! :-)
10:54:06matslLinusN: I know but would you prefer that I was just silent?
10:54:13eye69...and one of the very annoying bugs seems to be fixed.
10:54:33eye69Yep, fixed.
10:54:56bobTHChow many vote this day (without blank vote ;)) ?
10:54:57LinusNmatsl: let's hope Bagders vote script can handle blanks :-)
10:55:07Bagderit does
10:55:15eye69That one where you couldn't go to a specific tune in the playlist if shuffle was on...hated that
10:55:33BagderbobTHC: totally 34 voters atm
10:55:53Bagderah, yes that was fixed a while ago
10:56:07eye69Bagder: Have you guys been contacted by Archos themselves in any way? Have they expressed any opinions about your project?
10:56:23LinusNno, they have been silent
10:56:24Bagdereye69: not a word, not a beep, not a comment, nada, nothing ;-)
10:56:34eye69Bagder: That's bad.
10:56:40LinusNis it?
10:56:41Bagderis it really?
10:56:45bobTHCI'm sure only spying...
10:57:08Bagderwe've come so far now we don't need their help anyway
10:57:16eye69They'll probably try to steal the whole thing anyway. :)
10:58:00eye69They must feel like total dipshits...seeing what an OSS project can accomplish. :)
10:59:26 Join idefx [0] (
10:59:45eye69But of course, they sell more units...
10:59:59Bagderwe can't really be bad for them
11:00:32eye69My collegue just made up his mind on ordering one after he saw Rockbox on mine
11:02:14 Join gizz [0] (
11:05:32Mode"#rockbox +o Bagder " by ChanServ (ChanServ@services.)
11:05:49Topic"we NEED your VOTE!" by Bagder (
11:09:19Mode"#rockbox +o LinusN " by ChanServ (ChanServ@services.)
11:09:23Topic"Bagder wants YOUR vote!" by LinusN (
11:10:24eye69Vore on what?
11:10:27eye69Vote even
11:10:31LinusNt-shirt design
11:11:24gizzdone! :)
11:11:58*Bagder waits
11:12:41gizzbagder: any means for us to see vote stats ??
11:12:51Bagderonly when I leak the info ;*)
11:12:58eye69Btw, is ID3v2 supposed to be supported?
11:13:00Bagder6, 24 and 7 are the top-3
11:13:02LinusNgizz: no, no, no you must vote for Badger! :-)
11:13:05idefxHi, so, i've add 2 files to the apps dir (one .c, one .h), i've included the .h in main_menu.c, but when i try to make, i get an undefined reference. What should i do please ??
11:13:06Bagdereye69: yes
11:13:19eye69Bagder: It does not show v2...only v1
11:13:20LinusNidefx: make clean
11:13:30LinusNeye69: read the faq
11:13:38LinusNis your v2 tag large?
11:13:40Bagdereye69: try Q44 in the faq
11:13:41 Join alkorr [0] (
11:13:46LinusNhi alan
11:14:09idefxLinusN: already done.
11:14:58idefxLinusN: i'm not used to code big project, so, i don't really know how to work with that sort of things ;)
11:15:02LinusNwhat is the undefined reference?
11:15:16gizzlinus: remember my problem booting rockbox ?
11:15:17Bagderand do you build a simulator or for target
11:16:17gizzlinus: well, I could boot yesterday an archos .mod (disk booting vs ROM booting)
11:16:29idefxLinusN: undefined reference to dbtest (which is a function coded in the .c file i add, and refered by the .h file in the line: "extern int dbtest(void);"
11:16:41gizzlinus: but rockbox still doesn't boot :(
11:17:45LinusNidefx: simulator or target?
11:18:13idefxsimulator. Havent tested target yet
11:18:20LinusNgizz: windows?
11:18:30eye69Btw, Q12 @ FAQ has the old name for freenode
11:18:33LinusNidefx: then you have to add the file in uisimulator/Makefile
11:18:41gizzlinusN: what you mean ??
11:18:50LinusNdo you use windows?
11:19:01gizzyes, and linux too
11:19:11LinusNdid you copy the mod file using windows?
11:19:14idefxLinusN: Ok, and if I wan't to do it for target, have to add the file in ... ??
11:19:25LinusNidefx: nowhere, that is done automatically
11:19:44alkorrgizz: XP ?
11:19:49idefxLinusN: OK. Thks ;)
11:20:00LinusNgizz: make sure that the mod file doesn't have strange attributes, like read-only, system etc
11:20:05gizzalkorr: 2000
11:20:16gizzah yeah ?
11:20:28gizzI'll try that
11:20:29alkorrit doesn't matter
11:20:35LinusNwin2000 does silly things to speed up USB disconnection
11:21:16gizzarchos.mod (rockbox) has only archive set
11:21:20alkorrare you respectful of what you must to do to unplug USB ?
11:21:26gizz(it is generated by make)
11:21:33gizzalkorr: yes
11:21:41gizz(I use DevEject :)
11:21:47eye69Hmmm...guys...Q17 in the faq says "More cool features with the wire remote control (including controlling your Archos from your car radio (req hw mod))". What wire remote control would this be?
11:21:57alkorrhow do you rename your old archos.mod
11:22:03gizzanyway, I copy archos mod by the same means (and it boots)
11:22:08eye69alkorr: Just rename it.
11:22:28eye69alkorr: Preferably something like archos.mod.<version>
11:22:43gizznope, I cp over
11:22:43alkorrand it worked that way ?
11:22:51LinusNBrowse to the root folder of the Jukebox. Make sure you are viewing ALL files (including hidden, system and read-only files). Delete the ARCHOS.MOD file, and replace it with the one you want (5.03a I presume). Then right-click on the ARCHOS.MOD file and make sure the properties are NOT read only, NOT system, NOT archive, and NOT hidden.
11:23:22alkorrproperties don't matter
11:23:34alkorrthe naming matters
11:23:39LinusNIn order to quickly allow disconnects of USB media, Windows 2000 often does not properly re-index the FAT after marking some files as "erased". This rarely occurs in Windows 98/ME and should not occur in XP.
11:23:49gizzI think it's more about FAT32 part check..
11:24:16gizzremember my jukebox fell on the floor
11:24:19alkorrsometimes renaming in such way fools the stock firmware which thinks it is a valid archos.mod
11:24:57gizzI'm sure rockbox does a test more than archos does for the partition
11:25:10eye69LinusN: Is there a way of playing an entire dir, with subdirs, without making a playlist?
11:25:18Bagdereye69: no
11:25:27Bagdernot yet
11:26:39alkorrwell it could a ordering problem coupled with the renaming bug
11:27:12alkorrit did happen to me.
11:28:22gizzLinusN: what do you think ? Is my problem a rockbox bug, or a feature ? (if archos _can_ boot my (semi-dead)hdd, and rockbox _can't_)
11:28:50alkorrgizz : which doesn't boot ? archos or rockbox ?
11:28:55gizzLinusN: if it is a bug, I can continue tracing my bootups, to catch the bad check..
11:29:04gizzalkorr: rockbox
11:29:25gizzalkorr: it panicf() after logo when archos boots normally
11:29:41gizzalkorr: and MP3 can be still played
11:29:44alkorrdo you get the archos firmware or the archos software from HD ?
11:30:06gizzalkorr: I tried both, and both worked (ROM _and_ disk .mod)
11:30:16alkorrokay i misunderstood your problem
11:30:36alkorrin fact jukebox really read you rockbox software
11:31:07gizzand rockbox don't boots till the end
11:31:13alkorrtake WinHex
11:31:25Bagdergizz: I think it sounds as if rockbox fails on something that it shouldn't need to fail on
11:31:36gizzbagder: exactly
11:31:38alkorrmaybe the first boot is bad and as archos can read the other copy of boot it can continue
11:31:43gizzbagder: I think it's in fat_mount
11:31:51Bagdersounds possible
11:32:32alkorrdo you read sector 6 in rockbox ?
11:32:53gizzanyway, I can continue tracing this as I have the material to study.. Do you think it can be benefic for rockbox project ?
11:32:54alkorri'm speaking about FAT32 boot
11:32:57alkorrnot mbr
11:33:45alkorrsectors 6,7 and 8 are copies of sector 0,1 and 3 (FAT32 boot, FSINFO)
11:34:01gizzalkorr: AFAIK, part table is in sect 0, and it is valid
11:34:19alkorrnot the MBR
11:34:23gizzalkorr: my FAT32 part boot is at sect 63
11:34:35alkorri'm speakin with relative sector
11:34:39alkorrnot absolute sector
11:34:44gizzso ?
11:34:52alkorrabsolute sector would be 63, yes
11:35:13alkorrwell try to check if secto 63 and 63+6 has the same content
11:35:27gizzOk, thx, I'll try that
11:36:02alkorrbecause if your first FAT32 boot is bad, you can always fix it with 63+6
11:37:06gizzand that's why Windows & archos don't complain, by walking through sector copies, right ?
11:38:49alkorrit may be the problem you get or maybe not
11:39:05alkorrdo you get a message ?
11:39:12alkorran error message ?
11:39:31gizzNo FAT32 pa
11:39:38gizzon the 1st line
11:40:57gizzalkorr: sect 63 and 69 are identical :(
11:42:04gizzbut look: I just tried to read 1000 sectors from 0, and it fails
11:42:10gizzhdd rattles
11:42:16alkorrthis message is after a fat_mount ? you need to have gdb to get more verbose message...
11:42:23gizzwill try to see waht sector fails reading
11:42:32alkorroh bad
11:42:49gizzalkorr: I didn't make the serial patch, so I place lcd_puts to see msgs
11:43:07alkorrDEBUGF -> LCD ?
11:43:11idefxwhat's the max length of a filename (including fullpath) ??
11:43:32gizzalkorr: nope, my own msgs (rc printing, that stuff)
11:44:09alkorri must depart now i have a date today
11:44:13idefxBagder: OK, thanks.
11:44:23gizzalkorr: hey, it fails even by reading 100 sectors from 0
11:44:29alkorrtry with WinHex to see if you can read sectors
11:44:35alkorrunless you use Linux
11:44:54gizzbye all
11:45:08 Quit gizz ()
11:46:23idefxBagder: woups, what must i include to have MAX_PATH
11:47:21idefxBagder: Thanks ;)
12:07:19Bagderzagor's awake ;-)
12:08:06*Bagder runs for lunch
12:11:46 Part eye69
12:25:58***Saving seen data "./dancer.seen"
12:29:03 Join gizz [0] (
12:29:04 Nick LinusN is now known as LinusN|lunch (
12:29:17gizzalkorr: r u there ?
12:29:45HesSomeone remove from the mailing list, QUICK!
12:30:03Hesa mail is looping between his mail system and the mailing list
12:30:59Hesa really stupid mail filter which filters on a stupid trigger and sends a warning back to the mailing list and then filters it's own warning.
12:31:05Hesand sends another warning. Excellent!
12:37:34alkorryes i'm back
12:38:58alkorrgizz: solved ?
12:40:08gizzalkorr: ahh :)
12:40:31gizzalkorr: I found out that sector 61 is unreadable (an only this one)
12:40:51alkorryes ?
12:41:04alkorrdoes rockbox read sector 61 ?
12:41:16gizzbut why ?
12:41:23alkorrwere is the parameter sector ?
12:41:33gizzdo you think ata_read_sectors should skip bad sector ?
12:42:33alkorrthe problem does not come from ata_read_sectors but the reason why we need the sector 61
12:43:20gizzmaybe rockbox needn't, but read anyway
12:43:44gizzcollecting more sectors than needed at one time
12:44:09alkorrwhere is the code to save parameters on disk along with RTC ram ?
12:44:45gizzyou'r right I presume
12:45:07gizzdo you know if sector 60 is free for this purpose ?
12:46:16alkorrshould be
12:47:07gizzwell, maybe the best for me would be to use that instead of 61; BTW, there are some values in 60..
12:47:35alkorrjust try to read 60 and 62
12:47:50 Nick matsl is now known as matsl|away (
12:47:54alkorrto see if there are use as copies of parameter sectors
12:47:59gizzthey read ok
12:48:08alkorrtheir contents
12:48:47alkorryeah !!!! 61 is sector for parameters saving !
12:48:53alkorrin settings.c
12:49:35gizzoops, 62 contains some FAT32 boot sector :(
12:49:57gizzmaybe 60 could be used
12:49:58alkorrokay another copy
12:50:09alkorryes i don't think you have choice
12:50:21gizzBTW, it's you the ata guru ? :)
12:50:51alkorrit is quite unstill that your sector 61 cannot be read ? what happened ?
12:51:02alkorrwe can say that
12:51:14gizzwell done !
12:51:22gizzmy jukebox fell on the floor
12:51:34gizzI suspect the heads where near sector 61 :)
12:51:45alkorroh yeah, it must surely be that
12:51:55gizzone question : do you think it's possible to see when
12:52:14gizzplugged to USB, when the host unmount it ?
12:52:27gizzI mean, some ISD register is modified
12:52:34gizzah ?
12:52:43gizzothers said me no
12:53:03 Quit matsl|away ("Client Exiting")
12:53:04alkorryou must unplug USB
12:53:22gizzI thought, rockbox could detect when host unmount usb, but oh well..
12:53:24alkorrthere is no way to consult ISD register on the jukebox side
12:53:32gizzu sure ?
12:53:39alkorri'm sure
12:54:20alkorrISD can provide a I2C connection but is only to get some info
12:54:29 Join matsl [0] (
12:54:36alkorrnothing which could be interest for us
12:55:06alkorrbesides ISD I2C is not coonected on the I2C bus of jukebox
12:55:18gizzno fun
12:55:34 Quit matsl (Client Quit)
12:55:51alkorrthe only in/out pins you have is a USB enable pin and USB plugin detect pin
12:56:34alkorrUSB plugin detect can be linked with an interrupt to raise when USB is plugin.
12:56:39 Join matsl [0] (
12:56:55gizzand that's what is done.. right
12:56:55 Quit matsl (Client Quit)
12:57:41gizzwell, I'll try to use the sector 60 modif :)
12:57:46gizzbye all !!
12:57:49alkorrjust a minute
12:57:59alkorrtry with a stock archos firmware
12:58:16alkorrrecorder or player ?
12:58:16gizzyou mean, disk .mod ?
12:58:30gizzalready tried : it works
12:58:31alkorrok it must save its parameter on disk
12:58:47alkorrjust check when changing parameters
12:58:55alkorrand rebooting jukebox
12:59:10alkorrit still keeps the default value or reflect the new value
12:59:27gizzyou mean, discover where archos (if) store settings
12:59:30alkorrif changing is really saved, it means it can use another sector
12:59:54gizzok, but
13:00:11gizzwhat if archos uses an on disk hidden file ?
13:00:24alkorri'm doubtful
13:00:39alkorryou can check it with seeing hidden file
13:00:49gizzright, impossible
13:01:04gizzgood idea, I'll see that, bye !!!
13:01:32 Quit gizz ()
13:03:59 Nick LinusN|lunch is now known as LinusN (
13:05:46 Join matsl [0] (
13:06:57Bagderdamn naturesbest
13:07:11idefxisn't it ...
13:08:05idefxmatsl: am'working to the db stuff. May have lots of things to think about.
13:08:27matslNice to have things to do.
13:10:39idefxwell, the problem is the size of the file: it can be more than 2 meg long, so, we can't load it it ram as you can imagine... so, i wonder how we should do it ...
13:11:03BagderI still get the problem zagor fixed yday
13:11:11*Bagder dives into it
13:11:50Bagderbtw, I stopped the naturebest crap
13:11:56matslidefx: read a part at a time?
13:12:59idefxyes, sure, but how long should the part be ?? and, this mean we could forget the sort functionnality, except perhaps for small dir, or if we can sort the list itself (waiting for writing file possibility;)
13:13:25LinusNBagder: what problem?
13:13:30matsl/nick matsl|phone
13:13:38BagderLinusN: Zagor's wps.c 1.131 fix
13:13:50Bagderin which he ignores the KEYRELEASE if it wasn't pressed in the wps too
13:13:59Bagdertry this:
13:14:14Bagderin wps press f1 => menu press left to exit
13:14:18idefxwould you think a 1000 char array is too big ?
13:14:28Bagderthe left is then treated in the wps
13:14:52LinusNBagder: which version are you running?
13:14:58Bagderthis morning
13:15:01BagderI checked the code
13:15:05Bagderhis fix was ruined
13:15:08LinusNelinenbe broke it
13:15:28LinusNbut i think he fixed it again in 1.136
13:15:42Bagderah, right 5 hours ago
13:15:51Bagderwhich was after my version
13:16:01BagderI have 07:13
13:17:13matslidefx: (Back again) Well we should use the possibility to sort the file on the PC
13:17:52matslidefx: That's the major idea with creating a db after all isn't it.
13:18:52idefxYes. That's the best solution I think.
13:20:12matslidefx: With some positioning info in the db we could use lseek to jump to the interesting data. (But how efficient that is. I don't know.)
13:22:21idefxyes, sure. Don't know too. But, we should forget the idea of lining positionning, cause we have to scan each char to know if it's a line end. So it would be better to make a byte positionning system.
13:26:30alkorrbtw, gizz cannot read sector 61, which is parameter sector. Is it enough to cause a part error ?
13:27:11matslidefx: Hmm.. I guess jumping with lseek pays of best if you can jump over a sector or two.
13:28:07idefxfat sector you mean ?
13:28:32Bagderalkorr: I dunno
13:28:54idefxand how long are fat sectors ??? 4 Ko ??
13:29:03Bagder512 bytes
13:29:14idefxOK, 512 Bytes.
13:29:30idefxSo a 1024 char tab would be fine.
13:29:41alkorrno the sector is in the first track, where you can find MBR
13:29:49alkorrsector 63 is FAT32 boot sector
13:30:14alkorrsector 61 is used to save settings
13:30:28Bagderalkorr: they're not discussing that at all
13:31:12alkorrgizz told me its player feel down once and must have damaged this sector. No he has problem to run rockbox
13:32:31alkorrso my question is to know if you really need sector 61 to let rockbox run (it is for player)
13:33:55Bagderalkorr: check out the code
13:34:01alkorrso we have a really problem because we can easily damage vital sectors with leaving them on when having jukebox falling down.
13:34:07alkorryes but where ?
13:35:01BagderI'll dig a little...
13:38:11Bagderreads sector 61
13:38:32Bagderbut ignores errors
13:38:39alkorrjust a minute
13:38:45alkorrtell me if i'm wrong
13:39:17alkorryou just write sector 61 when reading or writing a sector ?
13:40:00Bagderwhen writing sector 61, we write sector 61, that's right
13:40:26alkorrhummm it changes nothing, error is not ack...
13:40:47alkorrno when you read any sector, you check a delayed sector
13:41:16Bagderit checks for delayed-write, yes
13:41:32alkorrbut anyway there is no reason rockbox writes this sector if we don't change settings...
13:41:53 Nick matsl is now known as matsl|busy (
13:42:00alkorrwhat I remembered it has a partition error dispalyed
13:43:31Bagderwhich indicates a problem with fat_mount() or something
13:45:11Bagderright, fat_mount() returns an error code
13:45:15alkorrThere a signature ROCK in this sector ?
13:45:31Bagderit doesn't reach there
13:46:18alkorrbecause if it can read the sector, variable 'correct' is false and 'load_config_buffer' returns error
13:46:44Bagdergizz said he got a text on the screen saying "no fat32 pa"
13:46:50Bagderthat is from apps/main.c
13:46:57Bagderbecause fat_main() returns error
13:47:11Bagderit does not involve the config reading
13:47:11alkorra message of its own because he has no serial output
13:48:58LinusNhehe, i just did a screen dump over the serial port from my recorder
13:49:21idefxLinusN: send it !
13:49:35LinusNthe bitmap?
13:50:08alkorrwhat happens if it cannot read settings secton in 'settings_load' , will it set a default one and activate delayed_sector ? so a any read attempts will try to write this sector at the end ?
13:50:21LinusNthat may be the problem
13:50:33Bagderwhy is that a problem?
13:51:06Bagderso it may fail to write it later, but that shouldn't cause problems
13:51:07idefxLinusN: yes
13:51:10LinusNdon't know :-)
13:51:28Bagderit fails on the mount
13:51:29Bagdernothing else
13:53:17alkorrok we fail to write a sector so ata_status is set for error
13:53:51alkorrwhen attempting to read a sector, we check the status for read ? only for ready or would accept error too ?
13:54:06 Nick seb-sleep is now known as seb-away (
13:54:14BagderI have no idea where you're heading with this alkorr ;-)
13:55:01alkorrtimeout because status has bit err set
13:55:19alkorrbefore issuing read command
13:55:34Bagderdidn't you look at gizz's problem?
13:55:37alkorri'm looking at 'wait_for _rdy' to check
13:57:18alkorrbefore reading partition, you're did a 'settings_load', so you try to read sector 61
13:57:47Bagderthat's done after
13:58:20alkorrsorry to say, it is not what i can read in main.c
13:58:21Bagderfirst fat_mounf() *then* read config
13:58:48Bagdersettings_load on line 150
13:58:49alkorrhumm yes
13:59:01alkorryou're right 'settings_reset' :/
13:59:15alkorrokay there is another problem too...
13:59:20LinusNfirst of all, it finds that the partition table is invalid
13:59:28LinusNsince it tries to mount(0)
13:59:38LinusNso the partition table is the key here
13:59:43LinusNsector 0
13:59:52LinusNwe need a dump of that
14:00:05alkorryes i see
14:00:20LinusNor rather, it fails to mount a fat32 system on the four partitions
14:00:32idefxlisten that and tell me what you think about it: for the db feature, make a db_struct containing 2 512 bytes buffer, one for current work, and the other if we acceed to the end of the first one. and when we use the db_struct, if one buffer is finish, then we use the second buffer and we re-fill the first one. is it a good idea ?? (i'm looking for something efficient, that doesn't consume much memory)
14:00:42LinusNso the partition boot sector may be interesting as well
14:00:49alkorrbut he told me that fat32 copies are same and it works with archos firmware and windows so...
14:01:00alkorryeah MBR
14:02:22alkorrdoes that mean archos use a copy of MBR. Windows can solve this problem by scanning possible FAT32 boots... well well mystere !
14:10:57alkorri'm looking for disk_int... where is it ? disk.c ? in which folder ?
14:11:33alkorrah common, okay
14:16:40 Join notch [0] (
14:17:15alkorri want to have DEBUG message (without GDB), using DEBUG in tools/configure would be good ?
14:22:30alkorrplease, it freezes on mine
14:23:40alkorrdoes that mean rockbox needs an ACK before issuing serial messages ???
14:24:13LinusNalkorr: DEBUGF talks gdb protocol
14:24:23LinusNit waits for t e'+'
14:24:29LinusNthe '+'
14:25:03alkorrbut nothing else ?
14:25:14LinusNit sends the data as hexascii
14:25:25LinusNreplace the function with a plain send loop
14:25:40alkorrit is not my question. Sorry i'll try to explain
14:25:48alkorrapart of debugf
14:25:59***Saving seen data "./dancer.seen"
14:26:17LinusNwe use the serial port for remote control only
14:26:24alkorris there any other gdb operation like init which could block because of a need of ACK ?
14:26:25LinusNthe transmitter isn't even enabled
14:26:50alkorrwith tools/configure I asked for DEBUG version
14:26:53LinusNevery debugf call wants an ack
14:27:06alkorryes but i can shut off '+4
14:27:07LinusNthe debug version links at 91000000
14:27:22LinusNbecause the gdb stub is at 9000000
14:27:32alkorrso the trouble
14:27:39LinusNthe debug version is only for gdb operation
14:28:00alkorri'm not sure to have a sh1-gdb...
14:28:20LinusNyou should hack debug.c yourself to enable serial debug output without gdb
14:28:23alkorrso I must hack...
14:28:52LinusNand be sure to commit that hack afterwards
14:34:29LinusNalkorr: take a look at my latest commit in drivers/serial.c
14:34:51 Nick edx|sleep is now known as edx (
14:34:52LinusNyou can use that sending function if you want
14:36:44alkorri want DEBUGF messages so i still need to hack debuf.c indeed
14:37:03LinusNyeah, but you don't have to reinvent the serial transmit
14:37:16LinusNi am silly
14:37:33LinusNthere already is a serial output function in debug.c
14:38:24alkorri'm not planning to reinvent the wheel but use those of debug.c ;)
14:39:27alkorrit would be interresting to have a special case to output message without gdb
14:39:49alkorrbut if it is only me who could do it well it is not important
14:39:50idefxjust a question: if I open the N first bytes of a file and put it in a char array, is the last byte of the array filled with '\0' ??
14:43:09LinusNalkorr: i would want a serial output without gdb as well
14:43:37idefxLinusN:no to me ??
14:44:04LinusNno, read() doesn't put anything in the buffer except what was in the file
14:44:14idefxLinusN: Thks
14:46:32 Join gizz [0] (
14:48:34gizzalkorr: the fault is not due to settings retrieval : it hangs before; problem is in fat_mount
14:49:23alkorryes i knwow, there was a discussion between me and Linus and Bagder
14:49:34gizza yeah ?
14:50:01alkorrand we finally think your bad sector 61 is not only the faulty one.
14:50:44alkorryou must check your MBR
14:51:16alkorrthat is, does it contain valid partition infos ?
14:51:24gizzI've done extended scandisk on it, but nothing appeared as a problem
14:51:41gizzunder linux, everything seems correct
14:51:56gizzone fat32 partition, accessible as /dev/hdc1
14:52:06gizzand disk is of cours /dev/hdc
14:52:22gizz(mounted as IDE of cours)
14:52:34alkorrin fact, we need a more precise message, where is the problem located ?
14:53:12alkorrMBR, one FAT32 ? why ? unreadable sector ? bad info in part info ? etc.
14:53:13gizzthe message is : No FAT32 pa
14:53:23gizzit's panicf()'d in main.c
14:53:45gizzwhen no part has been found
14:53:47alkorrnot enogh precise. It seems to be a personal message not in the current source
14:54:27alkorrwhere does this message lye ?
14:54:28Bagderpanicf("No FAT32 partition!");
14:55:04alkorrokay it does not find any valid partition...
14:55:11gizzlisten : I've investigated :
14:55:34gizzin main.c, when trying to find a valid part with fat_mount,
14:55:36 Nick elinenbe|sleep is now known as elinenbe (
14:55:52alkorrso chance for "pinfo[i].start" are not good
14:55:56elinenbeI was wondering if anyone could test the menu wrap on the player
14:56:26gizzelinenbe: will try when I can boot rockbox on mine :)
14:56:40gizzI saw recent modif on wrapping yes
14:56:56alkorrgoes in disk_int.c
14:57:01gizz(my patch didn't work so good ;)
14:57:01LinusNalkorr: it may be fat_mount() that fails to identify the FAT
14:57:05alkorrdisk.c : disk_init
14:57:20gizzfat_mount returns -1
14:57:27gizzwhy ? don't know yet
14:57:33alkorrit may or not but we must be sure
14:57:40alkorrso we need to check
14:57:45LinusNit could be bpb_is_valid
14:58:21LinusNit could be the countofclusters check
14:58:47LinusNor even a read error when reading the start sector
14:59:03alkorryou know what, Gizz. replace DEBUGF with panicf for disk_init and fat_mount
14:59:15LinusNgizz: change fat_mount() so it returns different values for the different errors
14:59:49gizzexactly what I've done for 1 microsec :)
14:59:56alkorror change DEGUGF into DEBUGFF and #define DEBUGFF to be panicf
15:00:10gizzno rc changed is better :)
15:00:47gizzin ~20 sec, I'll be fixed about that :)
15:02:06alkorrLinusN: for debug without gdb ? you've done that ? because I cannot commit (still some problem with SSH)
15:02:14gizzbingo! fat_mount returned -1, which means
15:02:50gizz DEBUGF( "fat_mount() - Couldn't read BPB (error code %d)\n",
15:03:07gizzhere we are
15:03:07alkorrread error ?
15:03:31gizzI've to print startsector
15:03:37gizzjust to be sure
15:03:40Bagderit is 0
15:03:49Bagdersince fat_mount(0) is what fails
15:04:15Bagderso reading sector 0 seems to be what fails
15:04:20gizz if (!(rc = fat_mount(pinfo[i].start)))
15:04:28alkorrhaving an incorrect startsector (pinfo[i].start)
15:04:31gizzthere is the first loop first
15:04:38gizzright :)
15:05:19alkorri told you to check disk_init first to see whether pinfo[i].start were valid
15:06:23alkorrbad boot signature for example
15:07:31alkorrshould be not that because disk_init returns okay i suppose
15:07:41gizzyes I presume too
15:07:51alkorr(no "disk: NULL")
15:08:37gizzalkorr: when fat_mount fails, startsector=63
15:08:46gizz(which is correct)
15:09:08gizzalkorr: BTW, I tested archos mod : you were right :)
15:09:26alkorrthat is ?
15:09:39gizzalkorr: archos use sector 60, as my 61 is unreadable; and yes, it remembers settings :)
15:09:46pimlottcwho decides when to release an official version, bjorn?
15:10:02LinusNbjörn has the final word
15:10:10Bagderwe let him think he has ;-)
15:10:11alkorrso stock archos firmware uses copies of settings sector
15:10:17pimlottccould I ask that you tag the releases in cvs
15:10:25Bagderwe do
15:10:25gizzalkorr: surely
15:10:45pimlottcI don't see any release tags
15:11:02 Quit RipnetUK (Read error: 104 (Connection reset by peer))
15:11:13gizzalkorr: I'll investigate my pb..
15:11:21Bagderpimlottc: look again
15:11:24LinusNpimlottc: they are called v1_3, v1_2 etc
15:11:34gizzanyone: somebody took care of the mailing list pingpong problem ??
15:11:35alkorryes, sector 63 has problem to be read...
15:11:41pimlottcI was looking for the word "release"
15:11:49Bagdergizz: yes, I fixed it
15:11:59alkorrbut rockbox never looks at the copy at 63+6
15:12:00gizzthx :)
15:12:08pimlottcmore descriptive tag wouldn't hurt :)
15:12:10alkorrwhereas archos and windows does it
15:12:15gizzwait : 63 _IS_ readable
15:12:22LinusNok, so we have work to do
15:12:26Bagderpimlottc: hardly anyone uses them anyway
15:12:49alkorrgizz : add 6 to startsector when fat_mounting
15:12:55alkorrjust to see what happens
15:13:02LinusNpimlottc: what is more descriptive than v1_3?
15:13:26LinusNgizz, don't!
15:13:48gizzwhy ?
15:13:54LinusNok, maybe if you change fat_mount(), but don't just add 6 to the parameter to fat_mount
15:14:03gizzwhy ?
15:14:13alkorryes in fat_mount should be enough
15:14:22gizzdifference ?
15:14:31LinusNbecause it will place the whole filesystem 6 sectors forward on the disk
15:14:33pimlottclinusn - I'm not saying it's inadequate, just that it is easy to miss
15:14:56alkorrgizz, in fact don't modify partinfo[i].start he means
15:15:06LinusNfat_mount assumes that the partition begins at that sector
15:15:14alkorryes right
15:15:28alkorrfat_mount does a lot of other job
15:15:46alkorrwhich must not be shifted for 6 sectors
15:16:17gizzI planned to add 6 in the calls to fat_mount, in main.c, in the loop
15:16:22alkorrso just when you ata_read_sectors the fat32 boot sector
15:16:38alkorrthat is in fat_mount
15:17:01LinusNgizz: that will NOT work
15:17:10gizzso what ?
15:17:19alkorrerr = ata_read_sectors(startsector+6,1,buf);
15:17:23alkorrthat's all
15:17:29alkorrin fat_mount
15:17:30LinusNadd 6 to the sector number in the call to ata_read_sectors in fat_mount()
15:18:06alkorrstartsector is also used to computed the start of FAT table
15:18:23alkorradding 6 will be catastrophic
15:19:26alkorrLinus, you don't use FSINFO, do you ? i can read nothing about FSINFO in fat.c
15:20:03LinusNi don't remember what that is
15:20:54gizzalkorr: err = ata_read_sectors(startsector+6,1,buf);
15:20:55alkorrFAT32 extension (sector just fater FAT32 boot)
15:21:02gizzalkorr: still fails
15:22:11alkorrinstead of scanning 0 in FAT entries to get new FAT entries for file, FSINFO tells us where to find the first free cluster and the number of free cluster
15:22:30alkorrcan be seen as a speedup
15:22:43alkorrsame message ?
15:22:55alkorrthat is failure of reading sector ?
15:23:16LinusNFSINFO isn't guaranteed to be correct
15:23:40alkorri just wanted to be sure whether you use it or not :)
15:23:43LinusNand why would we use FSINFO when we don't write to the file system?
15:23:50gizz(no msg just startsector=63, and readsector(63+6) failed)
15:23:58alkorrsure i know that too :)
15:24:23gizzalkorr : have to hangup, will come back later; anything I can test ??
15:24:38alkorrthe error code is -1 ?
15:25:03gizzwell, in cvs, all error leads to -1, but I changed them to -1, -2 and -3
15:25:13alkorrand you have -1 ?
15:25:36LinusNok, you should do the same return code check on ata_read_secors
15:25:40alkorrmaybe we need to retry or have a longer timeout
15:26:03LinusNyeah, change the wait_for_bsy() timeout to 10 seconds instead of 4
15:26:11gizzthat's what I thought at the begginning
15:26:23LinusNor even better, output the contents of ALT_STATUS
15:26:24 Join Tsd [0] (Tsd@
15:26:25gizz(letting time to my hdd to rattle ;)
15:26:46gizzhave to hangup, will come back later; anything I can test ??
15:26:46alkorrsince your sector 61 is damaged maybe sector 62, 63, ... 63+6 are a little bit damaged and need a longer wait
15:27:15LinusNwell, i have to go now
15:27:15gizzalkorr: nope, under cygwin, dd on these other sectors don't need extra time
15:27:19LinusNbye all!
15:27:23alkorrjust do it
15:27:25Bagderbye Linus
15:27:32 Part LinusN
15:27:43alkorr(longer timeout)
15:28:05gizzmay I adjust wait_for_bsy ?
15:28:23alkorrtimeout in wait_for_rdy and wait_for_bsy
15:28:34gizz4->10 ?
15:28:40alkorrfor example
15:28:50 Join Zagor [242] (
15:28:58alkorrif failure increase 20 , 50, 100
15:29:25alkorrit looks like you need to change your harddisk :/
15:29:49gizzI think too ;)
15:29:58gizzwill test that bye !
15:30:47 Quit gizz ()
15:37:49merwin|zZzZz40Mb in bad sectors
15:37:54 Nick merwin|zZzZz is now known as merin (
15:37:54DBUGEnqueued KICK merin
15:38:13alkorrvery bad news
15:39:09 Join Snorlax [0] (
15:39:29Bagderhi Snorlax
15:40:11Snorlaxanythin' new on Mike Dirnt(the bassist of Green Day)
15:40:25merinwhat aout him?
15:40:35Bagderno, but we know CmdrTaco runs Rockbox ;-)
15:41:22Bagderthe man behind slashdot
15:41:24SnorlaxDirnt has got a JB recorder!
15:41:40merinDirnt has a JB recorder? hahah
15:41:54 Join gizz [0] (
15:41:55SnorlaxPROBABLY ROCKBOX!! =)
15:42:08gizzalkorr : my _Hero_ !!!! :)
15:42:18Bagdergizz: works now?
15:42:35gizzYeaahhh !!! Rockbox now Rocks again ;)
15:42:43meringawd damnit, my hard drive just fried
15:42:54SnorlaxEHHHHH! DAMN!!
15:42:55Bagdergizz: those timeout changes did it then?
15:43:15gizzAll I had to do : change settings sector to 60, and Wait_for_**y timeaout from 4 to 10 !
15:43:37gizzI love you all !!
15:43:44BagderZagor: any comment on us enlarging those timeouts?
15:43:50CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
15:43:50*Bagder blushes
15:44:00alkorrmaybe we should have dynamic timeout
15:44:26alkorrusing a variable
15:44:34Snorlaxwhat happend with the maillist, has it gone BAZURK!!? =)
15:44:56Bagderwell, not the list
15:45:05Bagderbut some silly mail server
15:45:07alkorrhuh no timeout can be increased in fact
15:45:27alkorrsince quicker harddisk will answer quicker
15:45:27merin68 frickin' messages????
15:45:36gizzbye all, have to put again my games on my player's cvs code :)
15:45:49Snorlaxgames for the player!
15:45:58Bagderyes we want player-games ;-)
15:46:02gizzc u later
15:46:04 Quit gizz (Client Quit)
15:46:12Snorlaxgimme player games!
15:46:33ZagorBagder: which timeouts are we talking about?
15:46:41alkorrZagor: we should increase the timeout in ata.c for slower harddisk
15:46:47Snorlaxmy friends have gotten used to that rockbox rocks on my jb6k, now I need something new to impress with!!
15:46:53BagderZagor: <LinusN> yeah, change the wait_for_bsy() timeout to 10 seconds instead of 4
15:47:01Bagdergizz's boot problems
15:47:01alkorrGizz cannot rockbox its harddisk because ata_read_sectors fails to read
15:47:15alkorrwhereas archos and windows can read those sectors
15:47:37alkorr(its harddisk rattles)
15:47:43ZagorBagder: fine with me
15:48:32 Part elinenbe
15:49:01 Join kargatron [0] (
15:49:03alkorrwait_for_bsy and wait_for _rdy i think
15:49:16merinBagder: Someone (apparently not me right now) needs to fix the Custom WPS... it's requiring an LF after each line (even the last one)... shouldn't require one
15:49:28merinat least shouldn't require one on the last line i mean
15:49:51*merin is stupid... broke Track Skip :-)
15:50:10Bagderwell I repaired it ;-)
15:51:12merinI'm not sure why the lastbutton thing doesn't work on the player... for all intents and purposes it should
15:53:11merinhehe, sorry for the multi-line paste, but I had to show this off ;-)
15:53:12merinWindows has made corrections to the file system.
15:53:12merin 19,525,456 KB total disk space.
15:53:12merin 336 KB in 13 hidden files.
15:53:12merin 28,176 KB in 1,744 folders.
15:53:12***Alert Mode level 1
15:53:12merin 12,138,432 KB in 22,394 files.
15:53:14merin 40,784 KB in bad sectors.
15:53:16merin 7,317,712 KB are available.
15:53:18merin 16,384 bytes in each allocation unit.
15:53:20merin 1,220,341 total allocation units on disk.
15:53:22merin 457,357 allocation units available on disk.
15:53:24merin A disk error occurred during writing of the file allocation table.
15:53:34merinI think that's a record for me for shortest lasting HD
15:54:06 Nick merin is now known as merwin (
15:54:06DBUGEnqueued KICK merwin
15:54:06***Alert Mode level 2
15:54:11merwindamn, i can't even spell my own name today
15:55:03 Quit merwin ()
15:58:17kargatronas a rockbox user, nothing more, i find it difficult to parse the feature request tracker at sourceforge.
15:58:30kargatronlike, deciding whether a desired feature has already been requested.
15:58:40kargatronis there a search option?
15:59:21kargatroncan you sort on last-updated somehow? Does it do any good to add a note to an old request? Will developers see it?
15:59:40Bagderit is a bit crappy
15:59:59Bagderall notes get mailed to the rockbox-sf mailing list
16:00:04idefxkargatron: i've made a search function... But it's damn slow (45 sec to scan 64 files).
16:00:07Bagderwhich many developers subscribe to
16:00:07kargatronah, ok
16:00:31idefxkargatron : so i'm working on a database system to get the search function quicker.
16:00:52kargatronso a 'me too' (with extra detail) note will actually get seen (whether or not it's given any weight)
16:01:19Bagderyes, every change to every request will be seen by a number of developers
16:02:44kargatrondo developers choose to work on features they're most interested in, or are they parceled out by the group based on popularity/importance?
16:02:50 Nick matsl|busy is now known as matsl (
16:03:19Bagdermost of our development is based on what people feel like doing, more than doing smart choices
16:04:07***Alert Mode OFF
16:04:20 Join elinenbe [0] (
16:07:49 Join merwin|dizzy [0] (
16:08:46 Quit matsl ("Liece")
16:11:59 Join gizz [0] (
16:17:58Bagdermerwin|dizzy: try my wps commit I just made, should fix the wps problem
16:26:01***Saving seen data "./dancer.seen"
16:28:12 Nick merwin|dizzy is now known as merwin (
16:28:35merwinBagder: what commit was that? sorry, can't try much... HD is crapping out and trying to do data recovery before it goes all the way :)
16:28:59 Quit gizz (Read error: 60 (Operation timed out))
16:29:35merwinoh, the newline feature :)
16:32:19merwinso what's this screenshot feature?
16:32:48Bagderit can send the framebuffer to the serial port
16:33:06merwinso basically useless unless you've torn apart your archos?
16:33:07BagderI guess we'll use it for docs
16:33:13BagderI think so
16:33:48Bagderbut I might put it into the simulator
16:34:42merwinI still think we should bag the simulator idea, and just do a simple Emulator that only emlates the LCD instructions :) It'd be much less hassle in the long run.
16:35:08Bagdersure, you do that
16:35:16Bagderand I'll use the simulator until it works B)
16:35:23merwinif i knew how to write an emu i would :-)
16:35:40Bagderit's a huge work
16:35:52merwincan't be any huger than the project we've done so far!
16:36:16Bagderpossibly not, but don't expect the same crowd on such an undertaking
16:36:39alkorremu for what ?
16:36:46Bagderthe archos
16:37:08alkorrare you speaking about an SH1 emu ???
16:37:22merwindoesn't seem like it would take TOO long since all of the research has been done on all of the parts :)
16:37:23Bagderyes, plus the rest
16:37:39merwinBagder: no, only a limited SH1 emu and LCD emu.
16:38:00alkorrSH1 emulation is pretty forward. Which is most difficult it is its peripheral modules I think
16:38:15merwinthe LCD is the only part that we would need to really emulate
16:38:17alkorrwhat ... is ...
16:38:20merwinother than the SH1 i mean
16:38:52BagderI don't think it is worth all the work
16:39:19merwinThe player SIM doesn't even work in Win32 really
16:39:28 Part kargatron
16:39:38merwinThe recorder SIM works, but I can't exit out of MP3's :-)
16:39:44alkorrwe use DMA, serial, timer of SH1 which are peripheral modules. You need to be very close from the reality, which is not a simple task to do
16:39:51 Quit mecraw12 (Read error: 104 (Connection reset by peer))
16:40:20Bagdermerwin: so you say it is easier to write a new emu than to fix the sim?
16:40:43Bagderit works fine on x11
16:40:55alkorri did think about that possibility to be able to have a recorder on my pc (programming as if it is a true SH1, like BOCHS does for X86)
16:41:02merwinBagder: I never said easier at the moment... just less hassle in the long run.
16:41:13Bagderveeeery long then
16:41:49alkorrwell use SystemC ;)
16:41:50merwinfor someone who knows what they're doing (ie: has written an emu before), it shouldn't take more than a couple weeks
16:42:03BagderI beg to differ
16:42:04Bagdera lot
16:42:26Bagderbut I wouldn't mind one
16:42:46CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
16:42:46*merwin wishes he knew how to program an emu :)
16:43:06alkorrmerwin: write your SH1 emulator and we will see
16:43:30merwinalkorr: I can barely write a program to create MIDI files out of Nokia Ringtones, let alone an emulator
16:43:50merwinalthough I *did* do all of that in VB, which is a feat in itself
16:44:28alkorrhuh an emulator in VB ? it is not the answer i would like to hear :/
16:44:50merwinalkorr: I'm either fairly basic C/PHP, or advanced VB :-)
16:45:09merwinand Pascal if you put a gun to my head
16:45:57alkorri really dislikes VB...
16:46:14merwinalkorr: it's alot more powerful than you might think now... it's improved much since the 2.0 days
16:46:28merwinnative compiling is nice
16:46:35alkorri remember using it for Access, it slows down very much
16:47:03alkorrand have a very bad memory manager apparently
16:47:06merwinalkorr: nowadays, everything's done by activex, which is is used in VC and VB :P
16:47:39alkorryes... 'kro$oft
16:48:24*merwin is going to return his archos
16:49:14alkorryeahhhh I09;CPUaddr !
16:49:34merwinalkorr: from what?
16:49:58*idefx is happy, the archos db looks to work quite normally... perhaps even fine ;)
16:50:12merwinwhat is the archos db?
16:50:33alkorrfrom my source...
16:50:55 Nick merwin is now known as merwin|BestBuy (
16:50:59alkorrno problen not relative with the current CVSed version :)
16:51:37BagderI'm off
16:51:39 Part Bagder
16:52:10idefxwell, as it was far too slow to use the common id3 tools to get a nice search function, i made something based upon a kind of database created on the PC; it takes 30 seconds to "scan" (put the information in a struct, then display theym, and clear the display), of 1000 songs.
16:53:07merwin|BestBuyidefx: nice nice
16:54:35idefxit works with a 512 char buffer, and each db_entry -info one 1 song - take 300 char + MAX_PATH. that seems quite huge, but, as you only need one of each at the same time, it's not so much.
16:54:48idefxBut i have to go, i'll implement my search function later.
16:57:41elinenbeidefx: how will you implement text input?
16:59:00idefxelinenbe: that will be the last thing. If someone want's to do it, would be very kind ;)
16:59:26idefxelinenbe: wasn't it you who already made one text input methode ??
17:04:47adi|homeelinenbe: i was working on two ideas for it...
17:05:41idefxadi|home : could you send me what you did ??
17:07:53 Join RipnetUK [0] (
17:09:29adi|homeidefx i would.. but i scraped my first idea as on second thought it didn't make much sense...
17:09:45adi|homeand the second idea only has char layout on screen yet...
17:10:01adi|homeif you want i can catch you up to the brainstorming i was doing with ppl...
17:10:15adi|hometough.. that will have to wait till later.. i have to run into the office today :(
17:10:18adi|homewould that be cool?
17:11:37 Quit Zagor ("Client Exiting")
17:12:05alkorridefix : are you fond of Aterix and Obelix ?
17:19:09idefxadi|home : that'll be cool.. I have to go, sorry, my email is
17:19:55idefxalkorr: not at all (even if i live in britanny), but, i wanted something funny with "fx" inside (cause my name is François-Xavier)
17:26:45 Join LinusN [0] (
17:26:59LinusNWilma! I'm hooome!
17:27:36LinusNalkorr: any progress with the ATA problem?
17:28:11alkorrwell as you wanted, I move the transfering code outside the IRQ handler
17:28:21alkorrbut it fails with DISK: null
17:28:32alkorrso i'm disgusted :(((
17:29:18alkorri have now a wait_for_irq which return true when okay or false when error
17:29:40alkorrirq just set a ata_irq varaible
17:29:51LinusNok, but i was actually wondering about gizz's problem
17:29:58alkorrlet me send my last change
17:30:13alkorrbythe way it works
17:30:17alkorrfor Gizz
17:30:24LinusNwhat did you change?
17:30:26alkorrtimeout was problematic
17:30:30LinusNit was?
17:30:45alkorr10 instead of 4 was teh solution
17:30:55LinusNok, i should commit a change then
17:31:08LinusNi'll do that later this evening
17:31:33alkorr10 for timeout is not critical for quicker hd so it doesn't eat bread :)
17:31:44LinusNi am actually busy right now, i was just curious about gizz's problem
17:31:55LinusNbuilding Lego with my son
17:32:15LinusNcu later this evening when he has gone to bed
17:32:34 Part LinusN
17:39:00 Quit idefx (Read error: 104 (Connection reset by peer))
17:43:26 Nick edx is now known as edx|tea (
17:48:45 Join hardeep [0] (
17:52:29 Nick edx|tea is now known as edx (
17:53:24 Quit Snorlax ("gittar ny!")
18:01:36 Quit notch ()
18:05:17*merwin|BestBuy returns from his mission with a brand new archos
18:05:40 Nick merwin|BestBuy is now known as merwin (
18:11:27hardeepfinally converted to a recorder merwin?
18:14:14merwinunfortunately, no... they don't carry the recorders at best buy, so I can't upgrade to one quite yet
18:14:14 Quit mecraw (Read error: 104 (Connection reset by peer))
18:14:33merwinonce they do get recorders, then i'm going to upgrade
18:14:42hardeepdoh, that's right, a friend tried to get one there and couldn't
18:14:54hardeepeven though the guy at the store tried to convince him that it was the same laf
18:14:56 Join mecraw [0] (~lmarlow@
18:15:05merwinthey DO sell them on their online store, but i don't think that you can do returns through there
18:15:25hardeepdo they have fry's where you live?
18:15:30hardeep'cause they carry the recorders
18:16:19merwinyeah, i've been to fry's. The problem is, i don't want to purchase a whole new unit. If bestbuy starts carrying the recorders, then i can upgrade to it for almost free
18:16:27merwinNow Linus has his priorities straight:
18:16:27merwin<LinusN> i am actually busy right now, i was just curious about gizz's problem
18:16:27merwin<LinusN> building Lego with my son
18:17:10alkorrwell i'll think he's right
18:17:26alkorrby the way, it seems i solve my ata problem :)
18:17:32merwinalkorr: why?
18:17:38miahand at frys, you allways run the risk of buying something somebody has returned
18:18:32merwinmiah: ack, i don't want that
18:18:49merwinmiah: that's not legal, is it?
18:18:54alkorri'm changing the ata.c so it can use interrupt and multisector feature to speedup transfer (better use of hd cache) and reduce power
18:19:05hardeepthey put a sticker on it if it's been returned
18:19:15merwinalkorr: nice!
18:19:59alkorrfor the moment my jukebox read several files without complaining
18:20:10merwinalkorr: hopefully that will continue
18:20:12alkorri must be in the right way
18:23:15 Join Bill_k [0] (
18:23:38 Part Bill_k
18:23:54elinenbealkorr: way to go. sounds exciting
18:24:23merwinalkorr: any eta on a fully working patch?
18:25:54alkorreta ?
18:26:05merwinalkorr: estimated time of arrival
18:26:06***Saving seen data "./dancer.seen"
18:26:46merwinwow, very interesting... the new Jukebox is a Hitachi drive, not a Fujitsu
18:26:52alkorrwell I must check nothing is broken first and see with Linus if it is okay
18:27:40alkorrso maybe tonight. Since my SSH is broken, I think he would commit for me
18:29:49 Quit merwin ()
18:31:21 Join Silly_Fly [0] (luk@
18:32:14Silly_Flyisn't the format char for long %ld??
18:34:12 Join lohap [0] (
18:34:22 Join merwin [0] (~merwin@
18:34:40lohaphi all i was just wondering how do u use the wps config file and where does it need to be situated
18:35:27Silly_Flyisn't the format char for long %ld??
18:35:52merwinSilly_Fly: good ?
18:36:14lohapThe configuration file must be located in the /.rockbox folder of the
18:36:14lohapdevice and must be named default.wps
18:36:23merwinlohap: yup
18:36:23Silly_Flyi get ld instead of the variable...
18:36:28lohaponly problem in windows u cant start a folder name with a .
18:36:38merwinlohap: you have to use the command prompt
18:36:47lohapbarrr lol
18:37:03merwinlohap: need help with that part
18:37:03 Nick dwihno|gone is now known as dwihno (dwihno@Bald067.Baldakinen.Umea.SE)
18:37:07merwinhey dwi
18:37:09dwihnoGood evening, fellow ninjas
18:37:31lohapno its just an arse lol btw dosnt a . mean a hidden file
18:37:48Silly_Flymerwin : snprintf(lin[curline],sizeof(lin[curline]),"%ld",num[0]);
18:37:49Syntheonly on *nix
18:38:06 Quit hardeep ("Leaving")
18:38:13Silly_Flyand it displays ld (instead of the value of num[0])
18:38:15merwinlohap: it is also used for configuration files in unix
18:38:21merwinSilly_Fly: try just %d
18:38:23lohaphas anyone got a useful config fie for the recorder
18:38:40Silly_Flymerwin : it'll work with a long?
18:38:49Silly_Flymerwin : well, I'll give it a shot...........
18:38:50merwinSilly_Fly: i dunno
18:40:25Silly_Flywarning: int format, long int arg (arg 4)
18:41:20lohapon dirve i: which is my jukebox ihave created I:\.rockbox\wps.conf and the wps screen hasnt changed
18:41:52lohapsorry wps.config
18:42:46TsdIt have to be default.wps
18:43:31lohapyea just changed it and it works has anyone got any good wps configs for the jukebox
18:44:28alkorrit doesn't work when I remove debugf messages :(
18:44:45alkorrsorry, it wouldn't be for tonight
18:46:14bobTHCbye all
18:46:22merwinlohap: it needs to be called default.wps
18:46:22 Part bobTHC
18:46:57merwinFile Location
18:46:57merwinThe configuration file must be located in the /.rockbox folder of the
18:46:57DBUGEnqueued KICK merwin
18:46:57merwindevice and must be named default.wps (lowercase)
18:48:32lohapsortyed just playing with it now
18:50:48Tsdlohap: Mine is: %pr %pb %pt
18:50:54Tsd%s%ia - "%it" %fbkb %fv
18:53:18lohapthats only two lines i have 5 to play with
18:54:04TsdWell, I've got just AJB Player :\
18:54:53merwinme too
18:58:35lohapwell im sorted is there anythng else u can do using a .config file
18:59:31merwinyou can save multiple .wps files and load them using the dir browser
19:14:01 Nick Silly_Fly is now known as Silly_away (luk@
19:20:25 Quit lohap ()
19:21:02alkorrokay it seems to worl now...
19:21:26merwingood that it worls
19:22:41alkorri think there is a problem if we sleep mode or poweroff mode with ata
19:22:51alkorrthey flushed the cache
19:23:40alkorrso we get the worse situation when awakening for read transfers
19:24:00alkorrthat is maximal power consuption and spinup
19:24:51alkorrwas that issue debated about using or not sleep or poweroff mode ?
19:25:59alkorrjust a question, who uses a player ?
19:26:56alkorrwhen hd spins up, the scrolliing is disturbed ?
19:27:21Tsdwait a sec, need to test :)
19:27:49alkorri mean every time you get an access on hd while playing back
19:29:41Tsdyes, it keep scrolling
19:30:38 Quit merwin (Remote closed the connection)
19:30:39alkorrbecause i scrolling is interrupted when accessing (time depend whether data is already in cache)
19:31:29alkorrah yes i know why
19:31:41alkorri forget to add yield()
19:33:24 Nick Silly_away is now known as Silly_Fly (luk@
19:35:58alkorrit is better, even i can fell a slower scrolling
19:37:39alkorrwell would people be hurt with that ?
19:37:58alkorra slower scrolling when reading HD ?
19:40:03TsdI don't know, I can't see slower scrolling. It's normal :D
19:40:50TsdBut I use Player. May be it's different on Recorder
19:41:01alkorri change the ata.c so it can use interrupt and tansfer several sectors at once in one interrupt
19:41:32alkorrso it has some effect about the way the text scrolls
19:41:56 Quit sylvr (Remote closed the connection)
19:43:45elinenbealkorr: what is the effect of doing it like that? power changes? reliability?
19:44:17alkorrpower reduction by sleeping cpu
19:44:30alkorravoiding active polling
19:45:12alkorrmultisector feature so we can transfer several sectors instead of one per sector (so less polling too)
19:45:20elinenbeso, what do you think the increase of battery life will be?
19:45:30elinenbesounds good.
19:45:33alkorri don't know, it is a test
19:45:52elinenbedoes it give better reliability?
19:46:08elinenbewhat about for misreads (like jukebox is being shooken?)
19:46:10alkorrtoo soon to say
19:46:17elinenbeI see.
19:46:34alkorryou are spoken about error stats ?
19:47:29alkorrwell it is difficult to say, because using multisector helps to use better the HD cache
19:47:49alkorrso less power and faster transfer.
19:48:07elinenbeah... well, what about disk spindown feature?
19:48:23elinenbecan we go bleow 5 or 10 seconds now?
19:48:37alkorrspindown using sleep mode or poweroff mode is problematic, sure
19:49:11alkorrbecause you flush the cache and so increase the power
19:49:16elinenbeand we are not using that any more.
19:49:18alkorrwhen you read
19:49:36alkorri use standby mode
19:49:41alkorrfor the moment
19:50:44alkorri must discuss with Linus about the viability of using interrupt+multisector in ata.c to see if it brings some interest in our case
19:51:33alkorrof course it would be interesting to do some test to check the power consumption but i cannot do it
19:52:02alkorrall is theorical for the moment
19:52:07elinenbeuh huh.
19:52:09elinenbewell, good job.
19:52:14elinenbeI gotta go out.
19:52:28 Nick elinenbe is now known as elinenbe|skating (
19:58:01 Join Norrin [0] (
19:58:57 Part Norrin
20:01:40 Quit elinenbe|skating (Read error: 104 (Connection reset by peer))
20:15:35 Join g003y [0] (
20:17:33 Quit Silly_Fly (Read error: 110 (Connection timed out))
20:26:09***Saving seen data "./dancer.seen"
20:36:14 Quit juergen ("using sirc version 2.211+KSIRC/1.2.1")
20:37:27 Join gizz [0] (
20:43:18 Join Miran [0] (
20:48:53 Part g003y
20:54:35 Join sylvr [0] (
20:56:11gizzalkorr ? r u alive ?
20:57:44alkorri'm happy
20:58:26gizzwhy ?
20:58:41gizzabout my hdd problem..
20:59:04gizzI think the timeout increase did solve it, plus
20:59:22gizzI wonder if it didn't solve another problem :
20:59:25alkorrstill get a probklem ?
20:59:43alkorrdo you have a player ?
20:59:48gizzmusic skips, during running, roller skating..
21:00:26gizznow, it seems there's no skipping anymore (to be confirmed)
21:00:42alkorrwell i have a clean ata.c, could you test it ?
21:00:46gizzindeed, by skips, understand full STOP,
21:00:55gizzclean as what ?
21:01:04alkorrof course you must save the original
21:01:18gizzwhat is in it ?
21:01:38alkorri mean a clean ata.c which you can use with interrupt or not, with multisector or not
21:03:02alkorri modify ata.c, i explained to you there is a moment
21:04:02gizzbut how can I test it, if my problem is _already_ solved ?
21:04:11alkorrmultisector => several sectors at once instead of one per interrupt (something that ata use to tell us there is data to fetch)
21:04:28 Join merwin|dizzy [0] (~merwin@
21:04:32gizzwell, dcc it to me :)
21:04:35alkorrno i want you to test my ata.c to see what you think about
21:05:02alkorrabout music quality (no distortion, no skipping) because i'm not goot at heraring music
21:05:21gizzdcc it !
21:05:29alkorrok just a minute
21:06:17alkorrhey get DCC
21:06:54alkorrgizz, what do you wait for ?
21:07:11gizzhumm, remind how to get
21:07:19gizzI always send :)
21:07:32alkorr:/dcc get alkorr
21:07:32 Join elinenbe|skating [0] (
21:08:00gizzI've an error..
21:08:05gizzDCC Send from alkorr rejected (ata.c, file type ignored)
21:08:20alkorrlooks like you have a firewall
21:08:24 Nick merwin|dizzy is now known as merwin (~merwin@
21:08:41gizzok, mail it to plz :)
21:09:02elinenbe|skatingalkorr are you finished with the new ata.c?
21:09:06 Nick elinenbe|skating is now known as elinenbe (
21:09:37gizzbye all !
21:10:04merwinbye giz
21:10:11merwinelinenbe: it looks that way
21:10:16merwinelinenbe: skating?
21:10:26 Quit gizz ()
21:10:50alkorrit seems yes
21:10:59alkorrelin: it seems to work
21:11:33alkorri must do a test with multisector but interrupt
21:12:25alkorrhuh i mean without multisector
21:12:25elinenbemerwin: in-line skating... for exercise
21:13:00elinenbewhich would be better?
21:13:46Miranhave you managed to get the rockbox to stop skipping?
21:14:39alkorrwell increasing timeout apparently correct it
21:14:41 Join Zagor [0] (
21:15:00Miranis that for hd access?
21:15:09Zagorhi all
21:15:29alkorrelinenbe: you will have choice to set or not, i hope someone will check the power in different case
21:15:41alkorrZagor: good news
21:15:55alkorrmy ata.c code works with and without interrupt
21:15:56Zagoralkorr: what is that?
21:16:12alkorrwith multisector too
21:16:31alkorrnot tested yet with interrupt and without multisector
21:16:38elinenbealkoor can you send the ata.c to elinenbe at umich dot edu?
21:16:45Zagornice. do you see any performance improvement?
21:17:15alkorrspeed is not critical. We must check about the power indeed
21:17:45alkorrinterrupt helps us to sleep the cpu
21:17:58Zagorah, right
21:18:31alkorrmultisector reduces the time to transfer and then helps us to sleep sooner
21:18:57Zagorwe just need to actually sleep :-)
21:19:14alkorrthe only problem : because transfers takes several sectors, scroll text can slow down when transfering
21:19:45Zagorsend it to me, ok?
21:19:52alkorryield is not called at each sector but by indivisible group of sectors
21:20:07alkorri was waiting for Linus
21:20:38alkorrhuh your email is ?
21:22:00elinenbealkorr: me too! me too! elinenbe at
21:22:33alkorryou must define USE_INTERRUPT to use it and to set variable ata_multisector to 0 to disable multisector feature
21:24:40alkorrbtw the way i don't know if it works with recorder !
21:24:53alkorrbut shouldn't be a problem
21:25:00alkorrbtw i use standby mode
21:25:25merwinhey zagor
21:25:27alkorrZagor: is there a debate about using sleep mode or poweroff and the cache ?
21:25:40Zagoralkorr: not yet :-)
21:25:47alkorrbecause if we use sleep mode or poweroff the cache will be discarded
21:25:50Zagori haven't really told people about power-off yet
21:25:59Zagorsleep also?
21:26:13alkorryou need to reset the hardisk
21:26:21alkorrso chance the cache is reset too
21:26:37Zagorah right. well, we do 95% linear reads so the cache is not very important. power is more important.
21:26:46Zagormerwin: hi!
21:29:03alkorroh bad i leave led(on/off) in IRQ handler : to be removed
21:29:41alkorrotherwise you will feel like there is no hd access ;)
21:31:01elinenbeI was just skating with my player on the latest CVS, and there are still some bugs with it just jumping from the middle of one track to the start of the next track. Quite annoying.
21:31:10elinenbeIt happened at least twice.
21:31:19alkorrincrease the timeouts
21:31:21elinenbeand there was very little motion on the player.
21:31:36Zagorwe do still have bugs there, i know
21:31:42alkorrgizz told me he has no more skips since he increases the timeouts
21:31:58elinenbemy disk spindown is already 10 seconds.
21:32:04elinenbethat seems like quite a bit
21:32:21alkorri increase them to 10
21:32:31alkorrif not enought try 20
21:32:39merwinThe mailing list is turning into a conversation list :)
21:32:47elinenbeI was wondering if anyone here that has a player could try the menu wrapping?
21:33:06merwinelinenbe: i did try it, it works
21:33:08merwinelinenbe: thanks by the way
21:33:13alkorrin ata.c: wait_for_rdy, wait_for_bsy change 4 with 10
21:33:39elinenbemerwin: great!
21:33:45elinenbecheck this out: "Smart Power" NEX IIe™'s improved circuit design allows for drawing power from USB. This not only allows you to save battery life during song transfers but it can also power itself via AC power adapter.
21:33:56elinenbethat would be nice on the Archos.
21:34:27Zagorelinenbe: only 500mA 5V is allowed over USB and the controller alone draws 98mA. charging via USB is not gonna happen.
21:35:02alkorrelinenbe : i must update my sources first . what's the menu wrapping ? a scrolling ?
21:35:22merwinalkorr: you get to the bottom menu and it wraps back to the first one
21:35:39alkorrah ok
21:37:54 Nick seb-away is now known as seb-school (
21:38:17 Nick seb-school is now known as _seb_ (
21:39:01merwin_seb_: that was a quick trip to school
21:40:33elinenbeanother odd bug happened while I was skating. It stopped playing, and then when I looked at the screen of the recorder it was completly blank. I hit the OFF button, and then the left button to get back to the directory, then left again, and all I saw was the cursor and it seemed to have locked up.
21:41:12alkorroh something i got when trying to change ata.c. Looks like a read error
21:42:22alkorrthe next time you skate, don't forget to increase the timeout to check if problems still occur
21:45:02alkorri'm trying with sleep mode...
21:48:02alkorrhow much time i must set the spindown ?
21:48:11MiranIs there anyway to make the led screen, power down slowly, as par similar to the archos firmware?
21:48:39alkorrbacklight ?
21:48:47Miranthats the beast
21:49:01alkorra fade out ?
21:49:09alkorri tried it without success
21:49:29Miranhumm, it seems to work
21:49:36Miranthough, intermittently
21:49:37elinenbeZagor: you are a linux man. I have a question. I want to clone a 20 gig hard drive to a 240 gig RAID (2 120 gig drives) it is hardware RAID, so the controller will deal with the RAID stuff, but
21:49:41alkorrwith red led no problem using an interrupt but the same with backlight was a complete failure
21:50:03elinenbewhat I want to do is make a clean copy and increase the size of the partitions at the same time.
21:50:07elinenbeis that possible?
21:50:30alkorrthere is a capacitor which delays the state change so if you try to blink the backlight, well it doesn't work
21:50:53Miranit does :D
21:50:59Miranabout every other time
21:51:03alkorrwhich unit ?
21:51:07alkorrplayer ?
21:51:08MiranRecorder 10
21:51:18alkorrmaybe you could with a recorder
21:51:31Miranit seems to work already
21:51:52Miranmaybe the unit does it by default
21:52:07alkorrdon't say me that you use RTC to change the frequency of backlight ?
21:52:48MiranI wouldn't know, all I know, is that rockbox makes the backlight fade out
21:52:54alkorrwell sleep mode with interrupt and multisector works fine
21:52:57Zagorelinenbe: you do you want to clone it? if you change partition size, it's not a cloning
21:53:03Miranwhen it didn't use to on the 6gb recorder
21:53:46elinenbewell, I want to clone it and then increase the partition size
21:54:00alkorrseriously to fade out or in is impossible with baclight. Blinking it yes but not fade out on a player
21:54:05elinenbethe data is important, and I am NOT going to set eveything up from scratch
21:54:09Zagorwhy not a simple "cp -a /disk1 /disk2" ?
21:54:24elinenbethat should work okay?
21:54:35Miranah, on a player
21:54:36alkorrdd i=/dev/hd1 o=/dev/hd2 ?
21:54:42Zagorelinenbe: depends on what data it is
21:54:45Miranso it works on the recorder only
21:54:48Zagoralkorr: won't work with different disks
21:54:49elinenbeso, I boot off the old drive, set up the partitions on the new drive, and then copy
21:54:55elinenbethen set the new drive to be the master.
21:55:24alkorrmiran: i could do a great fade-in and fade-out with interrupt :)
21:55:37alkorrif recoder allows it
21:55:43elinenbealkorr: that would be awesome!
21:55:45Miranit does it anyway :)
21:55:49Miranwell... mine does
21:55:52Zagorelinenbe: cp does not preserve owners. if you want that, use tar.
21:55:55alkorrusin interrupt ?
21:56:00Miranthing is... old Recorder doesn't
21:56:02elinenbeug... what a pain.
21:56:08Miranout of the box, from stock
21:56:09elinenbeowners are very important here.
21:56:59alkorrbesause using very short interrupt helps not to disturb the normal execution, sou you coul be able to scroll a text AND fade in/out but you need to do so with a interrupt
21:57:41elinenbeI am looking for an elegant solution. I have to do it on 10 computers here, as it is a beowulf.
21:57:48alkorrMiran, give me your source
21:57:58Miranhum, I'm getting confused, I've never known a jukebox to have fade out on the backlight, 'cept this one
21:58:29MiranI had a recorder 6gb, it was very sick and was replaced yesterday with a recorder 10, this fades out
21:58:54Miranbut I thought the recorder 10 was practically the same
21:59:00alkorroh looks like a huge capacitor which discharges slowly...
21:59:49Miranhumm, could be, thing is.. the screen gets very bright before hand, then fades, as if more power is suddenly fed in
22:00:31alkorrRTC can output a frequency to blink the backlight. using a high frequency, i wonder if we divide by two the power drained by backlight ?
22:01:14alkorrdo you use this feature, Zagor ?
22:01:55Zagorno, we just turn it on or off
22:02:27Zagoroh, you mean the frequency? we actually use 2x the archos frequency
22:02:45Miranerm, is the wps supposed to disappear when it starts playing a song?
22:02:59ZagorMiran: define "dissapear"
22:03:16Miranits gone, there is no ID3 tag, no times, no progress bar
22:03:29Miranthe clock and battery are still there
22:03:38Zagorand what did you do?
22:04:03Miranthe recorder changed to the next track
22:04:11Miranthuh oh
22:04:35Miranat 0F0007FA
22:04:44Miranits died
22:05:07Miranharddisk light has started to blink rhymically
22:05:16Zagoryeah it does that to indicate error.
22:05:32Miranseems it doesn't like James Bond :D
22:07:20 Nick edx is now known as edx|sleep (
22:09:16Miranah there it goes
22:09:19Miranthe error is repeatable
22:09:27Zagorelinenbe: tar is your man. do like "tar cf - /disk1 | (cd /disk2: tar xpf -)" only you need some more options to save user etc
22:09:35MiranI think its just this song...
22:10:19 Quit merwin (Remote closed the connection)
22:10:36alkorrany succes with a spindown of 3s ?
22:12:18alkorrplease tell under which spindown limit you fail to have rockbox rocks fine ?
22:12:53Miranto try and counter the skipping
22:13:00alkorryou mean you need to set above 8 s ?
22:13:16MiranI find it skips when I walk with it
22:13:21Miranthe archos firmware is fine
22:13:26Miranbut rockbox.. skips :(
22:14:02alkorrtry to increase timeouts in wait_for_rdy/bsy, ata.c
22:14:07alkorr4 to 10
22:14:20alkorrans set spindown under 8s
22:14:37alkorrskipping or get next track ?
22:15:06Miranwith the spin down 3s, the rockbox still crashes
22:15:13Miranand the wps disappears
22:15:21 Join merwin|dizzy [0] (~merwin@
22:15:36alkorrrecorder ?
22:15:42 Nick merwin|dizzy is now known as recorder (~merwin@
22:15:44recorderalkorr: yes?
22:15:46Zagori'm checking in 10s timeouts now
22:15:49 Nick recorder is now known as merwin (~merwin@
22:16:10alkorrwho has a player ?
22:16:11 Nick merwin is now known as merwin- (~merwin@
22:16:14merwin-alkorr: me
22:16:15 Nick merwin- is now known as merwin (~merwin@
22:16:27alkorri'm at 3s spindown on my player
22:16:41alkorri didn't notice trouble
22:17:05alkorrmiran: recorder too ?
22:17:07Miranno skipping with 3s spindown?
22:17:12MiranI have a recorder
22:17:26Miranbut it skips with around 6-8s
22:17:33Miranas seen as the hd powers up
22:17:35Miranit dies
22:17:48alkorrusing poweroff mode ?
22:18:02Miranhow would I tell that to be on?
22:18:23ZagorMiran: are you compiling your own firmware?
22:18:41MiranAh, no, no I'm not
22:18:49Miranusing the builds
22:18:52Zagorright, i guessed not
22:19:08Zagortry the next bleeding edge, coming in ~5 minutes
22:19:22Miranwill do
22:19:42alkorrwell i try 1 s
22:20:10alkorrskipping are numerous ?
22:21:14Miraneverytime the hd powers up to refresh the memory, the unit stops
22:21:34alkorrbut the problem is i have a player not a recorder so maybe it works on player and not on recorder...
22:21:38Miranthe archos firmware used to stop, then after a while, recover
22:21:48Miranprobably :(
22:23:24alkorrzagor, i'm doubtful
22:23:46alkorri'm wondering if #define SLEEP_MODE really uses sleep mode...
22:23:50Zagoryeah, me too. but it's a good test.
22:24:19alkorrthere is no CMD_SLEEP_IMMEDIATE ?
22:24:42Zagoronly standby has a hardware timer
22:25:11alkorrokay... my hd still have vibration whereas i set 1s as spindown
22:25:58Zagorcan you hear it spinning?=
22:26:08alkorrto hear it no
22:26:12***Saving seen data "./dancer.seen"
22:26:18alkorrto touch it yes
22:26:31Zagorwhy would it vibrate it it's not spinning
22:26:45alkorrsleep mode shuts the motor off ?
22:26:51Zagornobody has reported this before, sounds very strange
22:26:56Zagoryes it does
22:28:40alkorrwell maybe the ata_sleep_mode fails but we don't see it
22:29:06alkorri should have a way to see when sleeping fails...
22:29:09Zagorwe would see it. the disk very much spins down for me, and everybody would scream at us otherwise
22:29:27Zagortry an old firmware
22:31:06MiranZagor: Is your lastest build, the 20:20?
22:31:17ZagorMiran: yes
22:31:18dwihnoZagor: I need a tip... Some kind of firewall/router single-floppy linux disk... :)
22:31:32Zagordwihno: to use for what?
22:31:42Miranbrb, I'm going to load the 2002-09-04 20:20:00 and take my recorder for a walk :)
22:31:44elinenbedwhino: smoothwall
22:31:47dwihnoZagor: since you'r the "Zagor knows stuff", I thought I'd ask you
22:31:54dwihnoelinenbe: what does it come with?
22:31:56CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:31:56*Miran Miran - Walking the rockbox
22:32:13elinenbeZagor: about the hard drive thing... I am looking for an elegant solution. I have to do it on 10 computers here, as it is a beowulf.
22:32:21elinenbedwhino: check out the web site.
22:32:26 Nick Miran is now known as Miran-Rockbox-wa (
22:32:37Zagorelinenbe: what is unelegant about tar?
22:32:55dwihnoelinenbe: you use it?
22:33:29 Quit Miran-Rockbox-wa (Read error: 104 (Connection reset by peer))
22:33:41elinenbedwhino: I have.
22:33:52elinenbeZagor: tar seems a hassle.
22:34:43elinenbeZagor: I would hope there was some command that would do it.
22:35:45Zagortar is the command. what do you want, clonediskbutwithdifferentsize?
22:36:04Zagoryou want a perfect copy of a file system, links users and all. tar provides that.
22:37:02Zagorsorry, didn't mean to sound harsh
22:38:22merwinHere's something interesting... My replacement archos player didn't have a fujitsu in it, it had a hitachi (i think it was hitachi at least)
22:38:29merwinit spins up really quick
22:38:44Zagormerwin: my rec20 had a hitachi too
22:39:00merwinZagor: is it just me, or when it's reading/writing, it sounds like it's going bad?
22:39:00Zagorclicks a lot, imho. toshiba is my favourite.
22:39:14Zagoryeah, it sounds a lot
22:39:18merwinZagor: OK, that clicking is what i was talking about... scared the crap out of me :)
22:39:33merwina 12V home charger is OK for the archos, right?
22:41:12 Quit RipnetUK ()
22:41:28Zagormerwin: for recorders, yes
22:42:21merwinZagor: how about player?
22:42:44Zagori'm not sure about player. it's a bit worse, charger-wise.
22:42:54Zagoryou have read the battery faq, right? :-)
22:43:20merwinZagor: yeah, but i thought i'd get a 2nd opinion... my charger has a 4.5v, 6v, 7.5v, 9v, and 12v setting...
22:43:34Zagorfor player, i'd go with 9V
22:44:02alkorrplug usb in when in menu results weird screen on my player
22:44:10alkorrnormal ?
22:44:13Zagoralkorr: yes, we don't handle that right
22:44:18Zagorwill fix soon
22:45:01merwinI wonder if the player stops charging when it gets full... nobody ever figured that out, huh?
22:45:07alkorrokay now i panicf when sleep fails
22:45:14Zagormerwin: it doesn't
22:45:40merwinZagor: that is kind of dumb :P
22:45:51Zagormerwin: indeed
22:46:01merwinZagor: why don't we have the player's battery status showing as charging when the charger is plugged in?
22:46:31Zagormerwin: i don't know. just nobody fixed yet.
22:46:47merwinZagor: any clue what the function for that is? probably pretty easy to add
22:46:54Zagorit's in status.c
22:47:02alkorrzagor: when occur the sleep mode; just 1 s after the last disk access ? when having 1 s for spindown
22:47:35Zagoralkorr: yes, but it also occurs immediately if no user activity and ata_sleep() is called. this is one of our problems.
22:47:49Zagorthus it can happen in the middle of an access, which is... bad :-)
22:48:26alkorrbecause when i call perform_sleep_... i should "panicf" if an error occurs (ret != 0) so...
22:48:45Zagorcase Q_SLEEP: should probably be "last_disk_activity = current_tick - sleep_timeout + (HZ/2);" instead.
22:49:07alkorrin ata.c ?
22:49:24Zagorbut try without that fix first
22:49:48alkorrwell i can hear music
22:49:56alkorrand still no panicf...
22:50:23alkorrso may be the disk has still residual vibration
22:50:48Zagoryou should be able to hear it spin down
22:50:56Zagorput it to your ear
22:51:37 Join RipnetUK [0] (
22:51:39 Quit RipnetUK (Remote closed the connection)
22:52:38alkorrokay i know, we need to reset to wake up. So I only need to forget it...
22:52:50alkorrwell i can hear the harddisk
22:52:59alkorrand some clicks when accessing
22:53:13alkorrbut still a drill
22:54:00 Nick dwihno is now known as dwihno|gone (dwihno@Bald067.Baldakinen.Umea.SE)
22:54:12 Join Miran [0] (
22:54:49MiranZagor: Recorder 10, with 20:20 firmware, skips, even with 3s, 6s and 10s
22:54:57Miranbut it takes longer to skip
22:55:07Miranthen 1.3 or other daily builds
22:55:37alkorronly timeouts are 10 ?
22:56:07ZagorMiran: ok
22:56:14Zagoralkorr: that was the only difference, yes
22:56:17alkorror are there other changes ?
22:56:59alkorrwell could you build for him with my ata.c (sleep mode, interrupt) to check any difference ?
22:57:17 Quit mecraw (Read error: 60 (Operation timed out))
22:57:20Zagorsleep mode?
22:57:28alkorr#define SLEEP_MODE
22:57:46Zagorwe always use sleep mode
22:57:56Zagorbut yeah, i'll make a test build
22:58:40 Join mecraw [0] (~lmarlow@
22:58:52ZagorMiran: can you receive dcc?
22:58:58Miranshould be ok
22:59:11alkorri temporaly remove the ata_reset in ata_read/write_sectors to see it would fails when sleep mode is active.
22:59:12Zagorok, sending a special test build
22:59:48alkorrdon't expect too much some improvements, Miran :)
22:59:50 Quit merwin ()
23:00:16Mirando you want me to see if it skips, yes?
23:00:50Zagorget some more excercise :-)
23:00:56alkorroh well i have disk: null so it sleep mode works
23:00:59Miranyeah, remove the fat
23:01:10Zagoralkorr: ok
23:01:21alkorri prefer that
23:01:34Miranok, which settings do you want me to try?
23:01:39Miranfor disk spindown
23:01:43ZagorMiran: 10s spindown
23:01:54alkorrthen decrease to 5, 3, 1
23:02:19Miranso hopefully, this should not skip :D
23:02:59alkorrMiran: are you sure your hd is ok ?
23:03:10MiranThe Recorder is new
23:03:16Miranand I formatted it yesterday
23:03:25Mirannice fonts Zagor
23:03:36pimlottcwhen the docs talk about some players only have 4 custom characters, is that the old pre-4.50 players?
23:04:27ZagorMiran: yeah, those are the proportional. we're not releasing them, going straight to loadable fonts instead
23:04:47Zagorpimlottc: yes. old have 4, new 8.
23:04:54MiranZagor: I see, they are very good :)
23:05:12 Nick Miran is now known as Miran-exercise (
23:05:25*pimlottc looks at the pre-4.50 charset
23:05:27*pimlottc sighs
23:05:42Zagorpimlottc: yeah, pretty limiting. thinking about games?
23:06:06Zagori'd like to see a simple bouncing dot on the player. should be pretty simple even with just the 4 chars on the old units.
23:06:11Miran-exerciseZagor: Cannot go below 3s for disk spindown, will do, 3s, 5s, 10s
23:06:16ZagorMiran-exercise: right
23:06:52Miran-exerciseZagor: Disk doesn't seem to spin down at all :)
23:07:14alkorrokay it is weird
23:07:31ZagorMiran-exercise: whoa, ok abort abort :-)
23:07:37 Nick Miran-exercise is now known as Miran (
23:07:56alkorrnormally it should spin down ?
23:08:09Miranit shouldn't be on ALL the time
23:08:52alkorryes but i never spins down or it does not abusely spin down
23:09:01alkorrit is what i want to understand
23:09:06alkorri = it
23:09:17Zagorit never spins down, even when idle. this is different with your code, alan
23:09:22Zagorjust tried it myself
23:09:46pimlottczagor - making a slidebar control for settings
23:09:50alkorrweird i didn't really touch your code about sleeping...
23:09:54Zagorpimlottc: ah
23:10:03Zagoralkorr: well, something is different :-)
23:10:31pimlottczagor - I have it working on my player but with 4 characters... will have to make some changes to say the least :P I assumed it would at least have square brackets, heh
23:10:36alkorryes but it is bad because we can use sleep mode... even without interrupt ?
23:10:54alkorri try it at once !
23:11:07Zagoryes we use sleep mode without interrupt
23:11:47 Join BoD[] [0] (
23:11:50BoD[]youhou !
23:12:07alkorrtu fais la fofolle ?
23:12:59 Quit mecraw (Read error: 60 (Operation timed out))
23:13:00BoD[]hein M
23:13:26pimlottcon the one hand, the accented characters in pre-45.50 are nice but the graphical characters are just wort of random
23:14:09 Join merwin [0] (~merwin@
23:14:23pimlottcwort = sort
23:14:45 Quit merwin (Client Quit)
23:20:10alkorrZagor: what does yield() ?
23:20:25Zagorstarts next thread
23:20:50alkorrbecause i call it when polling if an IRQ is occuring
23:21:19Zagoryes, that is the right way
23:21:35alkorri was wondering if transfer are delaying instead of being gather in a very short time
23:22:14alkorrpreventing the sleep mode to happen
23:24:07alkorrwhich priority has the ata_thread ?
23:24:35Zagorall threas have the same priority
23:25:02Zagorthis is yield-based, so no thread can have priority over another
23:25:22alkorrto explain rockbox never sleeps it shoulb be because it nevers calls ata_perform_sleep.
23:25:47Zagori will merge your code gradually and see where/if sleep breaks
23:26:18BoD[]what is this sleep thing ?
23:26:26ZagorBoD[]: disk spindown
23:28:15alkorrHZ is 1 s?
23:30:16alkorrwell with WinMerge, i will be fix
23:35:22alkorrwhat a silly i'm !!!!!!!!!!!!!!!!!!!!
23:35:38Zagorhehe, found it?
23:35:48alkorri put a return 0 in ata_perform_sleep !!!!
23:35:54BoD[]a haha :))
23:36:16Zagori know why it doesn't sleep... ;-)
23:36:33alkorrknew you mean
23:37:47BoD[]Who buys Lil Bow Wow's albums?? Seriously. Who.
23:39:59alkorrno vibration !!!
23:41:40alkorrweird it looks perfectly fluid
23:41:52 Join merwin [0] (~merwin@
23:41:53BoD[]hey slashdot is down ??
23:41:58alkorrwith 1s spindown
23:42:29alkorror maybe i should try another song...
23:43:15alkorrtransfer are really delayed because of yield()
23:43:37MiranBOD[]: doesn't work for me either :(
23:43:56alkorrbut it starts a little late
23:44:12 Join webbie_ [0] (
23:44:41 Quit webbie (Read error: 104 (Connection reset by peer))
23:45:54 Quit merwin ()
23:46:21alkorrZagor: can Miran make a test but with interrupt on (and return 0 removed) ? of course if Miran nods
23:46:49Miranbuild it for me and I'll see if it skips
23:47:43Miranwhat does the interrupt do?
23:48:02Zagortell us when to copy data from the disk
23:48:11MiranI see
23:49:22alkorrthe ata controller can send a signal to cpu to interrupt its execution and starts to handle its data
23:49:35alkorrso we call "interrupt"
23:49:50Mirandoes it do this, just as the data is about to run out, in memory?
23:49:54Miranor slighty before
23:50:29Zagorit has nothing to do with that. it's part of the ATA driver.
23:50:37alkorrwhen you ask for the ata controller to read data on disk, there is some time before you really get data
23:50:56MiranAhh I see
23:51:45Miranyou want the same again?
23:52:24Mirandoesn't go down to one
23:52:26alkorr1s seems to work in player at least
23:52:31Zagorhe can't go to 1, there's a limit
23:53:38 Nick Miran is now known as Miran-Exercise (
23:53:51Miran-ExerciseI'll be back in a while
23:54:02BoD[]i go to bed
23:54:04BoD[]see you
23:54:07BoD[]bye !
23:54:30 Quit BoD[] ("who")
23:54:54Miran-ExerciseZagor: Is the backlight suppose to light up when something in plugged into the 'Ear' socket?
23:55:26alkorri don't change anything about backlight
23:55:37Miran-Exerciseyou said that about sleep :)
23:55:52alkorri'm categorical here !
23:56:00ZagorMiran-Exercise: that's a bug in serial.c, i think. you get a VOL_DOWN button event
23:56:28Miran-ExerciseZagor: Ok
23:56:55Zagorit's just noise, but we are try to use everything we can! ;-)
23:57:15Miran-Exerciseheh, ok time to test this beast
23:57:23alkorrhey in fact i believed to be in standby mode, but in fact i was using nothing !

Previous day | Next day