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

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

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

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2015-01-23

00:00:18gevaertsKrist: tail just shows the last bit
00:02:23WaitwhatI mean I guess I have a Clip Zip floating around somewhere...but I'd rather have this big screen :(
00:03:00gevaertsKrist: "sdb: detected capacity change from 3933732864 to 0" isn't usual. I'd unplug and try again
00:04:11KristJust did that, fsck went through. If that doesn't allow OF to boot, I'll try the mkfs.msdos command
00:04:35CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:04:35*gevaerts nods
00:04:39KristJust to make sure I'm not messing up OF == orignal firmware?
00:07:19KristThat yields "Device Partition expected, not making filesystem on entire device, use -I to override." S\hould I use -I?
00:08:59KristReboot the device?
00:09:21gevaertsYes, into the OF. You'll have to reinstall rockbox
00:11:29KristWhen rockbox asks for the .bin file, the file it's looking for should be named "fuzpa.bin", right?
00:11:39KristWhen Rockbox Utility*
00:12:03gevaertsI think so, yes
00:12:22gevaertsBut that bit shouldn't be needed
00:12:29gevaertsThe bootloader should still be fine
00:13:31KristIt's still asking for it, and the only options are to load a file or abort
00:13:54 Quit ender` (Quit: A man without religion is like a fish without a bicycle.-- Vique's Law)
00:14:55KristLooks like that did it. It's running through the firmware upgrade now. I'll remember to actually eject/unmount my device in the future.
00:15:34[Franklin]yay \o/
00:15:45 Quit Krist (Remote host closed the connection)
00:17:14[Franklin]any way to cast (convert?) a fixed-point integer to a float?
00:17:37gevaertsThere's no good reason to want to do that :)
00:17:53[Franklin]I'm too lazy to convert /everything/ to fixed
00:17:59[Franklin]just the critical stuff
00:18:32wodzint i; float f = (float)i; easy
00:18:38 Quit TheLemonMan (Quit: leaving)
00:18:52[Franklin]wodz: with a nonzero number of fracbits
00:19:06gevaertsThen divide by 2^fracbits?
00:19:25wodz[Franklin]: you asked for casting fixed-point integer :-)
00:19:50[Franklin]heh yeah
00:20:08[Franklin]just saying "fixed-point" kind of implies a non-zero number of fractional bits ;)
00:20:17wodznot really
00:20:30[Franklin]to me, at least
00:22:00wodzanyway f = (float)i/(1<<fracbits) should work
00:22:35[Franklin]fixed-point also has the problem of reducing the maximum number I can represent...
00:22:41[Franklin]leading to divides by zero :/
00:23:29gevaertsWhat sort of numbers do you need?
00:23:32gevaerts32 bits is a lot...
00:24:21wodzWell, all games of my era used fixed point math. THIS IS DOABLE.
00:24:34[Franklin]of course
00:24:49*[Franklin] tries using 2 fracbits
00:26:04gevaertsThe thing with fixed point is that you need to know what your requirements are
00:26:52 Quit wodz (Quit: Leaving)
00:28:13 Join saratoga_ [0] (123e11e0@gateway/web/freenode/ip.
00:28:13 Quit saratoga_ (Client Quit)
00:34:32 Quit bertrik (Remote host closed the connection)
00:36:10[Franklin]I suppose I'll have to rework all the code to be resolution-independent
00:41:08 Quit AlexP (Remote host closed the connection)
00:48:13 Quit ZincAlloy (Quit: Leaving.)
00:49:02 Join ZincAlloy [0] (
00:50:16 Quit Waitwhat (Ping timeout: 245 seconds)
00:54:23 Join Waitwhat [0] (
01:15:33 Join williamtdr [0] (uid27909@gateway/web/
01:19:56***Saving seen data "./dancer.seen"
01:27:45 Quit RiD (Quit: A good plan today is better than a perfect plan tomorrow.)
01:50:41 Quit Waitwhat (Ping timeout: 245 seconds)
01:54:55 Join Waitwhat [0] (
01:55:52[Franklin]are there any preexisting fixed-point rounding functions?
01:55:59 Join micah [0] (~micah@debian/developer/micah)
01:56:34micahhey all, i got a sansa clip the other week and put rockbox on, went really smooth
01:56:55micahbut today I went to use it and it seems like its frozen... buttons wont move the menu, but I can turn it on and off
01:57:26micahi thought maybe it was locked, but it seems like I should hold the middle button and press play to unlock and that doesn't do anything
01:58:33micahit kinda feels like the down volume button doesn't work - it wont click like the up volume one, so I wonder if it got stuck somehow
01:58:40micahanyways, if anyone has any ideas of what to try...
02:06:30saratogaanyone noticed this problem:,50609.msg233742.html#msg233742
02:06:38saratogaclip zip poping noice
02:07:10saratogamicah: hold power button until it reboots
02:08:19micahsaratoga: i've done that, it comes back the same way
02:08:35saratogawhat do you mean?
02:09:16micahsaratoga: it shuts down when I hold the power button... but then when I press it again to turn it on, its still in the same state
02:09:48 Quit [Franklin] (Ping timeout: 245 seconds)
02:10:09 Join [Franklin] [0] (
02:11:26saratogadoes it work in the sandisk firmware?
02:14:38micahsaratoga: how do I switch to that?
02:14:47saratogahold left while booting
02:15:06micah... i had no idea
02:15:15micahRefreshing your media....
02:15:28 Join scorche [0] (~scorche@rockbox/administrator/scorche)
02:15:36micahsaratoga: no, its stuck there too
02:15:50micahi might be able to return it, but probably I'd need to get rockbox off
02:15:59micahor they are going to claim I did it
02:16:42micahthe home button works
02:16:48micahwhoa now it works
02:17:13[Franklin]have fun
02:17:16micahonly in the sansa firmware
02:17:34[Franklin]reboot into rockbox and try again maybe
02:17:48micahit wasn't working there. I pressed and held the home button, it locked, and then I pressed and held home again to unlock and it worked... but when I reboot to rockbox it doesn't work there
02:17:48saratogai thought it booted into rockbox?
02:18:34micahno, i booted it to sansa, it wasn't working there. then I pressed and held the home button, it locked, and then I pressed and held home again to unlock and it worked... then I rebooted to rockbox, and it doesn't work
02:18:52micahi'm wondering if its locked in rockbox and I'm just being an idiot about how to unlock it?
02:19:22saratogarebooting will unlock it
02:19:45micahweird, when i boot to sansa its locked again, but I can unlock it by holding down home (to lock it) and then holding down home to unlock it
02:20:05[Franklin]possibly a button's stuck?
02:20:18micahi think the down volume is stuck
02:20:37micahit doesn't click when I press it
02:20:45[Franklin]if all else fails, try taking the case apart and unsticking it
02:21:16micahthat will make it hard to RMA probably
02:21:48saratogayeah just send it back
02:22:25[Franklin]this leads me to wonder, what's the manufacturer's stance on rockbox?
02:22:36[Franklin](if you return a device with rockbox on it)
02:23:24micahhow do I remove rockbox and return it to the factory settings?
02:30:21 Quit Waitwhat (Ping timeout: 272 seconds)
02:52:08foolshmicah: flash a factory rom to it in recovery mode and deleting the .rockbox folder should place it back in a "like new" state as far as the system files go
03:00:23micahfoolsh: how do I get to recovery mode?
03:03:19micahwell, maybe doesn't matter, I flashed it while in the sansa firmware
03:06:29 Quit [Franklin] (Ping timeout: 252 seconds)
03:07:04 Join [Franklin] [0] (
03:07:13 Quit uwe_ (Ping timeout: 252 seconds)
03:09:12foolshmicah: I have had a few sansa products but not one of those, I just assummed it would have to the same type of flash proccedure, my bad :)
03:13:56 Join Strife89|Desktop [0] (
03:19:58***Saving seen data "./dancer.seen"
03:58:54 Quit [Franklin] (Remote host closed the connection)
04:15:22 Quit mc2739 (Ping timeout: 240 seconds)
04:17:31 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
04:24:15 Nick Strife89|Desktop is now known as Strife89 (
04:27:58 Join uwe_ [0] (
04:36:17 Join chrisb [0] (
04:51:36 Quit SpeakerToMeat (Ping timeout: 272 seconds)
04:58:04 Quit ZincAlloy (Quit: Leaving.)
05:20:00***Saving seen data "./dancer.seen"
05:39:21 Join Scromple [0] (~Simon@
05:45:17 Join ungali [0] (~ungali@unaffiliated/ungali)
05:47:16 Join SpeakerToMeat [0] (~SpeakerTo@prgmr/customer/SpeakerToMeat)
05:48:10 Quit Scromple (Read error: Connection reset by peer)
05:55:14 Join Scromple [0] (~Simon@
05:55:27 Quit TheSeven (Ping timeout: 265 seconds)
05:56:41 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:15:54 Join rela [0] (~x@pdpc/supporter/active/rela)
06:20:35 Quit rela (Ping timeout: 265 seconds)
06:32:10 Join rela [0] (~x@pdpc/supporter/active/rela)
06:37:01 Quit rela (Ping timeout: 265 seconds)
06:48:42 Join rela [0] (
06:48:51 Quit rela (Changing host)
06:48:51 Join rela [0] (~x@pdpc/supporter/active/rela)
06:53:27 Quit rela (Ping timeout: 265 seconds)
06:53:38 Quit Strife89 (Ping timeout: 246 seconds)
07:01:53 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
07:02:02 Quit dfkt (Disconnected by services)
07:02:17 Nick dfkt_ is now known as dfkt (dfkt@unaffiliated/dfkt)
07:04:43 Join rela [0] (~x@pdpc/supporter/active/rela)
07:09:40 Quit rela (Ping timeout: 272 seconds)
07:20:01***Saving seen data "./dancer.seen"
07:26:51 Join mortalis [0] (~kvirc@
07:37:14 Join rela [0] (~x@pdpc/supporter/active/rela)
07:42:16 Quit rela (Ping timeout: 265 seconds)
07:53:30 Join rela [0] (~x@pdpc/supporter/active/rela)
07:58:26 Quit rela (Ping timeout: 272 seconds)
08:04:01 Quit ungali (Ping timeout: 265 seconds)
08:26:00 Join rela [0] (~x@pdpc/supporter/active/rela)
08:30:36 Quit rela (Ping timeout: 265 seconds)
08:33:15 Join ender` [0] (
08:40:19 Join xorly [0] (
08:42:16 Join rela [0] (~x@pdpc/supporter/active/rela)
08:47:12 Quit rela (Ping timeout: 272 seconds)
08:51:40 Join petur [0] (5bb7304d@rockbox/developer/petur)
09:14:46 Join rela [0] (~x@pdpc/supporter/active/rela)
09:19:25 Quit rela (Ping timeout: 265 seconds)
09:20:02***Saving seen data "./dancer.seen"
09:22:03 Join AlexP [0] (~alex@rockbox/staff/AlexP)
09:31:25 Join edhelas [0] (~edhelas@
09:50:58 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:52:42 Join wodz [0] (
10:12:03 Quit xorly (Ping timeout: 264 seconds)
10:28:03 Quit pamaury (Ping timeout: 265 seconds)
11:01:42 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:10:44 Join pamaury_ [0] (~quassel@rockbox/developer/pamaury)
11:20:06***Saving seen data "./dancer.seen"
11:20:27 Quit pamaury_ (Ping timeout: 264 seconds)
11:22:13 Quit AlexP (Remote host closed the connection)
11:48:56wodzpamaury: I have weird qestion. Let suppose I have void (* const irqvector[])(void) array which holds IRQ handler pointers. Now some of this irqs are marked as reserved so I place UIRQ there. Now I'd like to UIRQ be of form void UIRQ(unsigned int) instead of void UIRQ(void).
11:50:02wodzpamaury: any idea how to hack this? One idea is to use inline asm in UIRQ() to grab the argument even if function prototype states it is void but I am not sure this will work actually.
12:40:35pamaurywodz: yeah that could work, as long as the caller for the irq handler puts the interrupt number in r0
12:41:11pamaurybut why not put the interrupt number in all IRQs ?
12:42:57wodzpamaury: That is an option, yes. It will look somewhat silly that I skip this parameter in all overwritten implementations.
12:43:55pamauryyeah that's true
12:44:17pamauryyou could probably get along with this using silly casts
13:09:17 Quit wodz (Quit: Leaving)
13:20:08***Saving seen data "./dancer.seen"
13:42:23 Join AlexP [0] (~alex@rockbox/staff/AlexP)
13:43:27 Join wodz [0] (
13:46:45 Nick suYin`OFF is now known as suYin (
14:12:45wodzpamaury: I came up with 2 versions Which one is less ugly?
14:23:50pamaurywodz: I think I prefer the first one
14:24:08pamaurybut it should come with a comment explaining the trick
14:25:46wodzpamaury: the first one, you mean with cast, right?
14:26:25pamaurywodz: I have a proposal for a third solution: in the assembly file handling the irq, write void __UIRQ(void )in assembly which does nothing but call void UIRQ(int nr)
14:26:48pamauryand in default_interrupt, put __UIRQ
14:29:15wodzright, this looks like the cleanest solution
14:29:24wodzthank you
14:42:06wodzpamaury: Something like this ?
14:43:53pamauryyeah, except I UIRQ should be __UIRQ and vice versa I think. Also I would write UIRQ in assembly (in crt0.S or whatever) or write it with the naked attribute
14:44:20pamaury(default_interrupt should alias to the void __UIRQ(void) version)
14:45:03wodzpamaury: naked attribute is ignored by mips gcc
14:45:21pamauryI see, then it's probably better to write in assembly don't you think ?
14:46:19wodzpamaury: since I don't allocate any registers the prologue and epilogue is empty. But yeah to be bullet proof I should put it in .S file
14:46:41pamaurywell atleast check what the compiler produces if you write it in C ^^
14:46:49wodzI checked
14:55:58 Join maruk1 [0] (
15:12:33wodzpamaury: The idea to move this into .S file doesn't work. The compiler barfs about aliasing function name to undefined symbol.
15:13:27pamaurywodz: you should probably define "extern void __UIRQ(void);" and in the assembly file make sure the symbol is exported (in ARM you use .global iirc)
15:13:45wodzeven though I put in .c file extern void __UIRQ(void); and in .S file I put .extern __UIRQ
15:13:58wodzerr .global __UIRQ
15:14:31pamaurywhat is the exact error ? maybe use objdump on the .o to double check
15:15:26wodzerror: ‘INT_MCA’ aliased to undefined symbol ‘__UIRQ’
15:17:02wodzin crt0.o I do have __UIRQ
15:19:00pamauryapparently (looking on google) this is a known/intential limitation of alias
15:19:56pamauryyou must define the alias in the same translation unit
15:20:10***No seen item changed, no save performed.
15:22:40wodzso the compromise is to leave trampoline function in .c file
15:23:56pamauryout of curiosity, did you try to alias the void bla(void) to void UIRQ(int) ? I'm not sure if alias checks for the type, I doubt it
15:23:58 Join amayer [0] (
15:24:45wodzit does not check the type :-)
15:24:52wodzI tried this
15:25:33pamauryso this could be a solution, just don't define a trampoline and alias directly to UIRQ(int)
15:29:18 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
15:30:38 Quit chrisb (Ping timeout: 272 seconds)
15:31:06 Quit fs-bluebot (Ping timeout: 265 seconds)
15:31:29 Join fs-bluebot [0] (
15:31:42 Join RiD [0] (
15:32:22 Quit bluebrother^ (Ping timeout: 240 seconds)
15:45:05 Quit AlexP (Remote host closed the connection)
15:49:23 Quit cmhobbs (Remote host closed the connection)
16:11:58 Join einhirn [0] (
16:19:03 Quit petur (Ping timeout: 246 seconds)
16:25:42wodzpamaury: Not really. Alias is only half of the problem, the other part is irqvector[] definition.
16:28:24 Quit krabador (Quit: Take the time.)
16:33:25 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
16:33:37wodzpamaury: Ha, there is yet another way:
16:34:23pamauryyeah I like that :)
16:35:11 Quit JdGordon (Ping timeout: 244 seconds)
16:35:53wodzwhat a mind twisting cheeting :P
16:37:24 Quit mortalis (Remote host closed the connection)
16:56:55 Quit JanC (Ping timeout: 264 seconds)
17:05:14 Join JdGordon [0] (
17:05:14 Quit JdGordon (Changing host)
17:05:14 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
17:05:21 Join booly-yam-4259 [0] (
17:06:08 Quit booly-yam-4259 (Remote host closed the connection)
17:07:00 Join booly-yam-4912 [0] (
17:07:31 Quit JdGordon_ (Ping timeout: 272 seconds)
17:10:04 Join JanC [0] (~janc@lugwv/member/JanC)
17:14:34 Join Misanthropos [0] (
17:16:28 Quit edhelas (Ping timeout: 265 seconds)
17:20:14***Saving seen data "./dancer.seen"
17:21:05 Quit maruk1 (Quit: Leaving.)
17:25:20 Quit wodz (Quit: Leaving)
17:25:40 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
17:33:26 Quit einhirn (Ping timeout: 246 seconds)
17:39:30 Quit Misanthropos (Quit: Ex-Chat)
18:08:57 Join AlexP [0] (~alex@rockbox/staff/AlexP)
18:46:11 Join booly-yam-4912_ [0] (
18:46:56 Quit booly-yam-4912 (Ping timeout: 244 seconds)
18:54:12 Quit renopt (Changing host)
18:54:12 Join renopt [0] (daem0n@unaffiliated/renopt)
19:10:58 Join lebellium [0] (
19:13:29 Join TheLemonMan [0] (~lemonboy@unaffiliated/thelemonman)
19:14:37 Quit pamaury (Remote host closed the connection)
19:20:16***Saving seen data "./dancer.seen"
19:20:51 Quit krnlyng (Ping timeout: 240 seconds)
19:24:09 Join krnlyng [0] (~liar@
19:54:57 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:03:16 Quit SpeakerToMeat (Ping timeout: 252 seconds)
20:05:54 Join SpeakerToMeat [0] (~SpeakerTo@prgmr/customer/SpeakerToMeat)
20:19:07 Quit KotH (Quit: beeeeep beeeep beeeep)
20:22:02 Join wodz [0] (
20:36:32 Quit booly-yam-4912_ (K-Lined)
20:40:24 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
20:45:50 Join ZincAlloy [0] (
20:45:58 Quit TheLemonMan (Remote host closed the connection)
21:20:18***Saving seen data "./dancer.seen"
21:30:29 Join naleo [0] (
21:32:26 Join naleo_ [0] (
21:32:29 Quit tchan (Ping timeout: 276 seconds)
21:32:38 Quit naleo (Client Quit)
21:33:02 Join krabador [0] (~krabador_@unaffiliated/krabador)
21:45:56 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
21:56:36 Quit naleo_ (Quit: Leaving)
21:57:19wodzAny particular reason we compile sound.c in bootloader build just to discard all functions it provides during linking?
22:12:38 Join booly-yam-6140 [0] (
22:19:00pamaurywodz: probably no one noticed
22:19:14pamauryadding more ifdef is not always for the best
22:20:49wodzpamaury: sure but it is really annoing when you try do provide the smalles possible subset of files to compile preliminary code during porting
22:21:11pamauryah right
22:21:17pamaurygood point
22:21:55wodzcompiling sound.c bring all codec defines and such
22:23:05pamaurywe should not compile it then :-p
22:27:36 Quit booly-yam-6140 (K-Lined)
22:42:58 Quit amayer (Quit: Leaving)
22:52:35 Join [Franklin] [0] (~franklin@unaffiliated/franklin)
22:58:51 Quit [Franklin] (Ping timeout: 240 seconds)
23:01:56 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
23:11:02 Quit stripwax (Quit: Miranda IM! Smaller, Faster, Easier.
23:17:58 Quit bzed (Remote host closed the connection)
23:18:08 Join bzed [0] (
23:20:22***Saving seen data "./dancer.seen"
23:45:29 Join ungali [0] (
23:45:30 Quit ungali (Changing host)
23:45:30 Join ungali [0] (~ungali@unaffiliated/ungali)

Previous day | Next day