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 2014-11-08

00:00:04foolsh_same behavior
00:01:44foolsh_But it seems to take longer to crash, let me track it through the functions, and confirm
00:07:15 Quit petur (Remote host closed the connection)
00:10:38foolsh_oh okay different bug this time
00:11:18*foolsh_ goes bug hunting
00:17:43[Franklin]foolsh_: progress!
00:17:52*foolsh_ nods
00:24:06 Join avatar [0] (
00:25:47avatarhey guys, I'm failing to register at the forum
00:26:24avatarWhat is 5 + 8? (enter wombat in the field):
00:26:32avatarAny guesses what the answer is?
00:27:22avatar(runs away hiding) I read it as "wombat in the field". Nevermind :)
00:27:23[Franklin]enter wombat!
00:27:32foolsh_avatar is a bot!!!
00:29:09foolsh_[Franklin]: "vm_init" line "rb->memset(m->vmVariables, 0, sizeof(m->vmVariables));"
00:29:23[Franklin]this should be easy
00:31:05 Quit avatar (Quit: leaving)
00:31:22[Franklin]foolsh_: weird
00:33:22[Franklin]foolsh_: perhaps try replacing memset with a for loop that zeroes all the variables?
00:39:55[Franklin]foolsh_: are you sure that's where the crash is?
00:40:05[Franklin]IMO it's pretty unlikely
00:40:50foolsh_I'll double check
00:41:06[Franklin]just 512B
00:41:42[Franklin]though if it is, that's OK because the audiobuf is *limitless*
00:46:39foolsh_okay yeah you're right, my fault, it down deeper
00:49:47[Franklin]foolsh_: where?
00:50:53[Franklin]any idea what line?
00:51:40foolsh_working on that
00:52:03*[Franklin] bets is either player_stop or mixer_stopAll
00:52:25foolsh_I raise you ten birds in the air :)
00:52:35[Franklin]actually, no , res_setupPart
00:57:55foolsh_nope not in res_setupPart
00:58:03[Franklin]then where?
00:59:25*[Franklin] now bets it's the memsets
01:00:04foolsh_rb->memset((uint8_t *)m->threadsData, 0xFF, sizeof(m->threadsData));
01:00:25[Franklin]knew it!
01:01:32[Franklin]there seem to be a lot of bugs involving memset :)
01:03:54[Franklin]well, I'm stumped
01:04:07[Franklin]you could try rewriting memset into a for-loop and see if that helps
01:13:53 Quit [Franklin] (Ping timeout: 265 seconds)
01:22:03***Saving seen data "./dancer.seen"
02:07:12 Quit bertrik (Quit: Lost terminal)
02:07:20 Quit TheLemonMan (Remote host closed the connection)
02:21:17 Quit uwe_ (Ping timeout: 244 seconds)
02:21:17 Quit Xyem (Ping timeout: 244 seconds)
02:21:19 Quit Guest66077 (Ping timeout: 244 seconds)
02:21:19 Quit Aldem (Ping timeout: 244 seconds)
02:21:19 Quit Mir (Ping timeout: 244 seconds)
02:21:34 Join Guest66077 [0] (
02:21:57 Join Aldem2 [0] (~Aldem@
02:21:59 Join Xyem_ [0] (
02:21:59 Join Mir_ [0] (
02:22:03 Nick Mir_ is now known as Mir (
02:22:23 Join uwe_ [0] (
02:23:07 Nick Aldem2 is now known as Aldem (~Aldem@
02:23:23 Quit Aldem (Changing host)
02:23:23 Join Aldem [0] (~Aldem@unaffiliated/aldem)
02:24:23 Part Aldem ("Leaving")
02:37:21 Join JdGordon [0] (
02:38:09 Quit JdGordon_ (Ping timeout: 250 seconds)
02:39:54 Quit uwe_ (Ping timeout: 244 seconds)
02:40:37 Join onder` [0] (
02:40:51 Join uwe_ [0] (
02:46:12 Quit onder` (Ping timeout: 265 seconds)
03:00:01 Quit AlexP (Remote host closed the connection)
03:22:04***Saving seen data "./dancer.seen"
03:53:41 Quit ZincAlloy (Quit: Leaving.)
04:11:38 Join onder` [0] (
04:46:32 Quit amiconn (Disconnected by services)
04:46:32 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:46:35 Quit pixelma (Disconnected by services)
04:46:36 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:46:37 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:46:38 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:47:21 Quit onder` (Ping timeout: 255 seconds)
04:49:24 Join onder` [0] (
05:22:05***Saving seen data "./dancer.seen"
05:46:23 Quit Strife89 (Ping timeout: 255 seconds)
05:56:24 Quit TheSeven (Disconnected by services)
05:56:36 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
06:15:15 Join rela [0] (~x@pdpc/supporter/active/rela)
06:20:04 Quit rela (Ping timeout: 258 seconds)
06:31:33 Join rela [0] (~x@pdpc/supporter/active/rela)
06:32:17 Quit prof_wolfff (Ping timeout: 255 seconds)
06:36:10 Quit rela (Ping timeout: 258 seconds)
06:47:38 Join rela [0] (~x@pdpc/supporter/active/rela)
06:52:03 Quit rela (Ping timeout: 244 seconds)
07:20:02 Join rela [0] (~x@pdpc/supporter/active/rela)
07:22:08***Saving seen data "./dancer.seen"
07:25:05 Quit rela (Ping timeout: 265 seconds)
07:26:32 Quit Provel (Ping timeout: 245 seconds)
07:27:42 Join Provel [0] (
07:36:16 Join rela [0] (~x@pdpc/supporter/active/rela)
07:40:57 Quit rela (Ping timeout: 258 seconds)
07:49:44 Join onder`_ [0] (
07:50:45 Quit onder` (Ping timeout: 264 seconds)
07:50:52 Nick onder`_ is now known as onder` (
08:01:07 Join AlexP [0] (~alex@rockbox/staff/AlexP)
08:08:39 Join rela [0] (~x@pdpc/supporter/active/rela)
08:09:30 Join onder`_ [0] (
08:10:51 Quit onder` (Ping timeout: 256 seconds)
08:10:56 Nick onder`_ is now known as onder` (
08:13:25 Quit rela (Ping timeout: 265 seconds)
08:14:55 Join onder`_ [0] (
08:17:30 Quit onder` (Ping timeout: 255 seconds)
08:17:33 Nick onder`_ is now known as onder` (
08:24:53 Join rela [0] (
08:24:55 Quit rela (Changing host)
08:24:55 Join rela [0] (~x@pdpc/supporter/active/rela)
08:29:44 Quit rela (Ping timeout: 255 seconds)
08:57:16 Join rela [0] (~x@pdpc/supporter/active/rela)
09:02:03 Quit rela (Ping timeout: 255 seconds)
09:07:52 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:13:30 Join rela [0] (~x@pdpc/supporter/active/rela)
09:18:19 Quit rela (Ping timeout: 258 seconds)
09:22:12***Saving seen data "./dancer.seen"
09:29:39 Join rela [0] (~x@pdpc/supporter/active/rela)
09:31:37 Join ender` [0] (
09:34:32 Quit rela (Ping timeout: 255 seconds)
09:48:11 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
09:49:27 Quit pamaury (Ping timeout: 256 seconds)
09:52:50 Quit stripwax (Read error: Connection reset by peer)
09:58:37 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
10:02:02 Join rela [0] (~x@pdpc/supporter/active/rela)
10:03:17 Quit stripwax (Client Quit)
10:06:29 Quit rela (Ping timeout: 255 seconds)
10:18:16 Join rela [0] (~x@pdpc/supporter/active/rela)
10:23:06 Quit rela (Ping timeout: 258 seconds)
10:24:45 Quit hob (Quit: meh :/)
10:50:40 Join rela [0] (~x@pdpc/supporter/active/rela)
10:54:53 Quit rela (Ping timeout: 244 seconds)
11:06:11 Join xorly [0] (
11:06:52 Join rela [0] (~x@pdpc/supporter/active/rela)
11:11:46 Quit rela (Ping timeout: 265 seconds)
11:22:14***Saving seen data "./dancer.seen"
11:22:36 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
11:28:55 Quit xorly (Quit: I quit, that is all)
11:36:43 Join xorly [0] (
11:39:14 Join rela [0] (~x@pdpc/supporter/active/rela)
11:43:41 Quit rela (Ping timeout: 255 seconds)
11:55:27 Join rela [0] (~x@pdpc/supporter/active/rela)
12:00:15 Quit rela (Ping timeout: 255 seconds)
12:30:26 Quit GNUtoo-irssi (Read error: Connection reset by peer)
13:01:12 Join JdGordon_ [0] (
13:02:26 Quit JdGordon (Ping timeout: 255 seconds)
13:22:03 Join TheLemonMan [0] (~lemonboy@unaffiliated/thelemonman)
13:22:16***Saving seen data "./dancer.seen"
13:27:39 Quit onder` (Ping timeout: 265 seconds)
13:33:14 Join hob [0] (~hob@
13:43:11 Quit lycide (Ping timeout: 245 seconds)
13:44:20 Quit K1773R (Ping timeout: 260 seconds)
13:44:43 Join K1773R [0] (~K1773R@unaffiliated/k1773r)
13:47:08 Join lycide [0] (~kvieta@
14:32:48 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
14:35:35 Quit bertrik (Ping timeout: 255 seconds)
14:38:49 Join onder` [0] (
14:40:12 Quit bertrik_ (Read error: Connection reset by peer)
14:40:31 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
14:46:16 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
15:00:23 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
15:13:08 Join lebellium [0] (
15:13:09 Join franklin [0] (
15:13:15 Nick franklin is now known as [Franklin] (
15:14:47[Franklin]foolsh_: o/
15:15:11 Nick foolsh_ is now known as foolsh (
15:16:14[Franklin]foolsh: I have no idea why it's crashing
15:16:25 Quit bertrik (Ping timeout: 256 seconds)
15:16:34 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
15:17:57[Franklin]256 bytes shouldn't be too much for the stack
15:18:34foolshTime for some gdb goodness
15:18:52foolshI have to work it out first though ;)
15:20:32foolshin the mean time I've been hording up ideas
15:21:49[Franklin]smells like feature requests to me :P
15:22:18***Saving seen data "./dancer.seen"
15:23:04foolshon target gdb? its been done before and there is a serial connection to use
15:23:57[Franklin]OneOfZombies is Ruby
15:25:36foolshYeah, I didn't hate on any languages, some aren't C that doesn't mean they're impossible though, just needs reimplemented in c
15:26:46foolshI was looking for simple but engaging prospects
15:27:02 Join ZincAlloy [0] (
15:28:09[Franklin]It's possible, but not that easy
15:29:42foolshI have also a list of "collections" of sources most in C
15:29:43 Quit [Franklin] (Read error: Connection reset by peer)
15:30:15 Join franklin [0] (
15:30:15 Nick franklin is now known as [Franklin] (
15:30:21foolshwelcome back
15:30:21[Franklin]argh kernel panic
15:31:22foolshRight before you left I sent "I have also a list of "collections" of sources most in C" don't know if you got it
15:31:49[Franklin]I did
15:31:58[Franklin]but then it crashed
15:33:13[Franklin]I like SCUMM the best
15:33:18 Join armisius [0] (ba2616c7@gateway/web/freenode/ip.
15:33:49foolshI think if one engine at a time was ported it would be best, there are quite a few
15:33:51[Franklin]armisius: hello!
15:33:58[Franklin]foolsh: yeah, of course
15:34:14foolsh[Franklin]: but controls are big hurdle
15:35:00foolsh[Franklin]: but perhaps with the point and click titles something can be worked out
15:35:09armisiusi have an digital photo frame and I was wondering if I could install rockbox on it, does any one knows?
15:35:11[Franklin]"Portability is a design goal"
15:35:28[Franklin]armisius: is it listed on
15:36:07foolsharmisius: Only if you do all the work, and have the skill and knowledge to do it I'm afraid, but what's the point? can it play sound?
15:36:22armisiusand it doesn't have a name
15:36:39[Franklin]armisius: so no
15:37:15foolsharmisius: also there is the requirements to run rockbox to consider, a photo frame is really very simple in it's capabilities
15:37:34armisiusok, another question
15:38:19armisiusthe photo frame have a dpfmate
15:38:27[Franklin]a what?
15:38:42 Quit stripwax (Quit: Miranda IM! Smaller, Faster, Easier.
15:38:45armisiusit's a software to download photos to the digital frame
15:39:31[Franklin]foolsh: ScummVM is C++, and it'd be pretty hard to convert it by hand like we did for xworld
15:41:49[Franklin]I'm thinking that I could automate at least part of the conversion process
15:42:10*foolsh would'nt mind starting smaller so he can catch up with [Franklin] :-P
15:42:21armisiusok thanks Franklin
15:46:45 Quit armisius (Ping timeout: 246 seconds)
15:46:58[Franklin]foolsh: I'm working on a cpp2c program right now :)
15:55:41 Join Strife89 [0] (
16:08:26 Join petur [0] (~petur@rockbox/developer/petur)
16:08:39 Quit [Franklin] (Ping timeout: 255 seconds)
16:42:17 Quit Strife89 (Ping timeout: 250 seconds)
16:43:15 Join polemon_ [0] (
16:53:54 Quit bertrik (Ping timeout: 250 seconds)
16:57:04 Quit Riviera (K-Lined)
17:20:45 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
17:22:20***Saving seen data "./dancer.seen"
17:28:29 Quit bertrik (Read error: Connection reset by peer)
17:28:47 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
17:34:05 Join krabador [0] (~krabador@unaffiliated/krabador)
17:52:27 Join franklin [0] (
17:52:28 Nick franklin is now known as [Franklin] (
17:52:32[Franklin]foolsh: back
17:52:39[Franklin]sorry about that
17:53:28 Join kugel [0] (~kugel@rockbox/developer/kugel)
17:59:25 Quit dfkt (Remote host closed the connection)
18:03:24 Join Riviera__ [0] (Riviera@2a03:b0c0:1:d0::10:b001)
18:13:09 Nick Riviera__ is now known as Riviera (Riviera@2a03:b0c0:1:d0::10:b001)
18:20:50[Franklin]foolsh: try the new patch set
18:21:01[Franklin]it replaces the memset call with a loop that does the same thing
18:21:05foolsh[Franklin]: ok give me a few minuntes
18:21:17foolshor minutes
18:25:15[Franklin]the VM struct isn't even a kilobyte
18:25:21[Franklin]I doubt that it's a memory issue
18:29:42 Quit krabador (Quit: Sto andando via)
18:30:44*foolsh thinks about a i386 port for his little eepc
18:30:57foolshit's really low spec
18:41:45foolshit crashes but I'm not sure if its the same place
18:44:42 Quit kugel (Ping timeout: 265 seconds)
18:47:13[Franklin]foolsh: it's in res_readEntries
18:48:36 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:49:41 Quit K1773R (Ping timeout: 244 seconds)
18:50:52 Quit bluebrother (Disconnected by services)
18:50:58 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
18:52:27 Quit TheLemonMan (Remote host closed the connection)
18:53:01 Join dfkt [0] (dfkt@unaffiliated/dfkt)
18:53:50[Franklin]which is called before vm_initForPart
18:54:01[Franklin]so the crash is happening earlier
18:54:19 Quit fs-bluebot (Ping timeout: 255 seconds)
18:54:49 Join trampel [0] (
18:57:00[Franklin] this is weird
18:59:21foolshI'm lost, let me go back over this
19:01:10 Join fs-bluebot [0] (
19:03:45 Join K1773R [0] (~K1773R@unaffiliated/k1773r)
19:05:26 Quit kugel (Ping timeout: 265 seconds)
19:11:08 Join onder`_ [0] (
19:11:42 Quit onder` (Ping timeout: 244 seconds)
19:11:47 Nick onder`_ is now known as onder` (
19:22:21***Saving seen data "./dancer.seen"
19:29:02[Franklin]foolsh: any ideas?
19:29:26foolshstill stepping through it
19:29:49foolshI really need to get gdb working
19:35:55foolshit's in vm_initForPart I swear
19:36:28 Quit bertrik (Ping timeout: 245 seconds)
19:36:31 Part trampel ("Leaving")
19:37:04 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
19:37:55[Franklin]foolsh: you sure?
19:38:17foolshI'm digging deeper
19:38:24[Franklin]what line?
19:38:36foolshthats what Im diggging into
19:38:40foolshhang on
19:47:17 Quit K1773R (Ping timeout: 256 seconds)
19:47:23foolsh[Franklin]: You'll never believe where
19:47:39[Franklin]tell me!!!
19:47:43foolshin your for loop
19:47:58foolshthe on you just wrote
19:48:19[Franklin]what line?
19:50:00foolsh this is enough to crash the player
19:50:54foolshstack over flow?
19:52:15 Join K1773R [0] (~K1773R@unaffiliated/k1773r)
19:53:21[Franklin]can GDB show stack usage?
19:53:56foolshyes it should
19:59:49foolshthats a very broad question
20:00:43foolshIt has been done before
20:02:01[Franklin]sorry, I can't
20:02:02[Franklin]no idea
20:02:49foolshdon't worry I work it out here
20:04:01foolshBut I suspect the "WTF?" I see in the comments is an indication of bad omens
20:06:30[Franklin]I am pretty sure that it is a stack overflow
20:07:50[Franklin]no, it's just weird
20:08:02[Franklin]it's modifying VM vars at runtime
20:08:13 Join prof_wolfff [0] (
20:08:13[Franklin]which is weird
20:08:21 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:08:52[Franklin]I do suspect that its a stack overflow
20:10:58 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
20:14:33 Quit bertrik (Ping timeout: 265 seconds)
20:18:37[Franklin]foolsh: I guess I'll use the audio buffer for the whole vm struct
20:18:53foolshsure why not
20:20:21[Franklin]probably the video struct too
20:21:04foolshlol can't hurt
20:22:51 Quit bertrik_ (Ping timeout: 250 seconds)
20:23:10 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
20:26:05foolshgeez I commented out everything in vm_initForPart except "rb->memset((uint8_t *)m->threadsData, 0xFF, sizeof(m->threadsData));" and it still crashes
20:26:16[Franklin]lol then its memory
20:28:55[Franklin]ok try the latest patch set
20:30:25[Franklin]if it still crashes, I might just allocate everything in the audiobuf :)
20:30:42[Franklin]it only needs about a megabyte total
20:36:04[Franklin]can you debug again?
20:36:38*[Franklin] will rofl when the audiobuf runs out :)
20:56:29[Franklin]foolsh: does it still crash in the for loop?
20:57:08foolshoh no it's much earlier, a few more minutes will tell
21:02:38[Franklin]ahh bug
21:02:38[Franklin]trying again
21:03:04[Franklin]try this patch set
21:03:23[Franklin]it's using a wrong chunk of the audiobuf for VM
21:04:08[Franklin]now it's using the right chunk
21:06:53 Quit mc2739 (Ping timeout: 245 seconds)
21:07:43foolshit crashed, but there is a different behavior now
21:08:00foolshit fills the bottem of the screen with a pattern
21:08:08foolshthen crashes
21:09:01 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
21:09:21foolshplugin_start Point 0 is the only output from logf
21:09:44foolshstill early, in engine_init I believe
21:10:55foolshengine_create I mean
21:12:00[Franklin]foolsh: hmm... bad memory access somewhere, I guess
21:12:13[Franklin]any more output?
21:12:32foolshI narrowing it down
21:12:56[Franklin]it wouldn't crash that early, I don't think
21:13:49[Franklin]well, I think ScummVM is a bit too hard
21:13:57[Franklin]perhaps Wolfenstein 3D should be next
21:14:07[Franklin]there's a nice SDL port:
21:14:22[Franklin]it is technically C++, but it should work in C just fine
21:14:56foolshthats what we thought last time :-/
21:15:11[Franklin]for what? xworld?
21:15:20[Franklin]just SDL that will be hard to port
21:15:24foolshOk do it does work
21:15:25[Franklin]it's not abstracted as well as xworld
21:15:31foolshon the SIM
21:15:32 Quit AlexP (Remote host closed the connection)
21:16:16foolshis the culprit now
21:16:24foolshlet me double check
21:16:30[Franklin]foolsh: hmm... really?
21:16:44[Franklin]I doubt it
21:16:45 Join AlexP [0] (~alex@rockbox/staff/AlexP)
21:16:55[Franklin]all it does is stop audio playback
21:17:01[Franklin]and get the audiobuf
21:17:04[Franklin]it doesn't access it
21:17:33foolshI'm only the messenger
21:18:02[Franklin]but yeah, double check
21:18:50foolshyes it's in there some where, let me dig in more
21:22:23***Saving seen data "./dancer.seen"
21:22:54*[Franklin] has his money on the plugin_get_audio_buffer() line :D
21:27:13foolshyep "res->_memPtrStart=rb->plugin_get_audio_buffer(&sz);"
21:27:46*foolsh goes to stretch his legs
21:30:40[Franklin]knew it
21:33:07*[Franklin] is stumped, again
21:44:32*foolsh shrugs "at this point I'm clueless"
21:44:45foolshsorry I'm not much help
21:45:25*[Franklin] is guessing that plugin_get_audio_buffer() is crashing
21:45:41*[Franklin] consults gevaerts the wise
21:47:11foolsh[Franklin]: so what happened to your ipod?
21:47:25*gevaerts hasn't seen any bug reports about other plugins that use the audio buffer recently
21:47:25foolshdid you break it playing xrick? :)
21:51:57 Quit [Franklin] (Ping timeout: 264 seconds)
21:52:08 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
21:53:19 Join [Franklin] [0] (
21:58:45 Quit pamaury (Ping timeout: 256 seconds)
22:01:57 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
22:02:16 Quit bertrik (Ping timeout: 264 seconds)
22:47:21[Franklin]foolsh: I'm really stumped :(
22:48:07foolshgive it rest then
22:48:37foolshnobody is paying you ;)
22:53:23[Franklin]foolsh: I'm guessing it's plugin_get_audio_buffer() that's crashing
22:53:35[Franklin]nothing's wrong with the left-hand side of the statement
22:57:00[Franklin] so yeah, it's the right-hand side
22:58:05[Franklin]gevaerts: any idea what's wrong?
22:58:12gevaertsThat's a fairly strong statement, given the number of other plugins that call plugin_get_audio_buffer() without issues
22:59:09[Franklin]then why does:
22:59:52gevaertsHow much stuff do you do before that?
22:59:58[Franklin]stopping audio
23:00:00gevaertsMaybe stack overflow?
23:00:09[Franklin]yes, that's been considered
23:00:39[Franklin]well, I'll check again
23:00:50[Franklin]but I find it pretty unlikely
23:01:05[Franklin]but then again, who knows? :)
23:02:22[Franklin]gevaerts: how would I used GDB to see how much of the stack is being used?
23:02:47foolsh[Franklin]: I'm afraid that would only work on target
23:02:54foolshnot in the sim
23:02:59 Join krabador [0] (~krabador@unaffiliated/krabador)
23:03:03[Franklin]foolsh: no, but sim should still use same amount of memory
23:03:10[Franklin](assuming 32-bit target)
23:03:39[Franklin]so all I need to know is how much of the stack is being used
23:04:01[Franklin][Saint]: ???
23:04:12[Franklin]ahh yes, sim will use memory differently
23:04:20[Franklin]but it would still be a good estimate
23:04:33[Franklin]nope. nope. nope.
23:04:39foolshand nope
23:04:51[Saint]behind the scenes, the two are wildly different, and one can't tell you anything particularly meaningful about the other.
23:05:02foolshthe sim is a sdl app mostly
23:05:14[Saint]that's why its a simulator, and not an emulator.
23:06:40gevaertsIt's a different CPU architecture, with different compiler versions and possibly different optimisation settings
23:06:50gevaertsSo stack usage might vary a lot
23:07:12[Saint]I thinkw e can safely substitute might with will.
23:08:03gevaertsWhat are sizeof(struct Engine) and sizeof(struct System)?
23:08:40[Franklin]well, how do I figure out the memory usage of a program?
23:08:40[Franklin]top won't cut it, there's no malloc
23:09:40[Franklin]finding out...
23:09:41gevaertsBecause those two structs are the only possibly big things on the stack by the time you reach plugin_get_audio_buffer()
23:10:42[Franklin]struct Engine is 6K
23:10:59[Franklin]System is 728B
23:11:04[Franklin]ooh that's a problem
23:11:18gevaerts6K is a lot
23:11:25[Franklin]perhaps just allocate all of Engine in audiobuf?
23:13:02gevaertsMaybe, or just as a global variable in the plugin buffer if you still have some space left there
23:13:07[Franklin]yep... that's it
23:16:35[Franklin]hmm... let's try it then
23:17:52[Franklin]ok foolsh try the latest patch set
23:18:09[Franklin]first it tries allocating it in the plugin buffer
23:18:16[Franklin]if its not big enough, it tries the stack
23:18:42foolshokey dokey
23:18:42gevaertsIf the plugin buffer isn't big enough, you almost certainly have a smaller stack
23:18:59[Franklin]true, but its worth a try
23:21:05gevaerts[Franklin]: just put it statically as a global variable...
23:21:24gevaertsIf the plugin buffer is big enough, that works, if not, the *linker* tells you
23:21:48[Saint]If the plugin buffer isn't big enough, it tries something _smaller_?
23:21:53[Saint]Sorry...but, what the fuck?
23:22:07*foolsh hits ctrl C
23:22:11gevaertsDon't go for complicated solutions if simple ones will do :)
23:22:24 Nick bertrik_ is now known as bertrik (~quassel@rockbox/developer/bertrik)
23:22:26***Saving seen data "./dancer.seen"
23:22:34[Saint]and check if your complicated ones even make sense first. ;p
23:24:05[Franklin]ok, now try
23:25:01 Quit rasher (Remote host closed the connection)
23:25:08 Join rasher [0] (~rasher@rockbox/developer/rasher)
23:26:16[Franklin]struct Resource is also quite large
23:26:16[Franklin](it makes up more than half of struct Engine)
23:26:16[Franklin]followed by struct Video
23:26:18[Franklin]at 2K
23:26:24[Franklin](Resource was 4K)
23:26:32[Franklin]and that's basically it
23:27:14gevaertsYes, but that's inside Engine, isn't it?
23:27:57[Franklin]so now its all off the stack
23:28:03[Franklin]foolsh: does it work?
23:28:25foolshhold your horses, I'm building clean
23:28:45[Franklin](if it doesn't work, I guess it'll all go in the audiobuf)
23:29:09*[Franklin] did a build in 15 seconds with his new, shiny, 24-core machine :)
23:29:59[Franklin]24 *PHYSICAL* core
23:30:36foolsh[Franklin] give gevaerts an big wet kiss on the mouth
23:30:59foolshAww but then take it back
23:31:15foolshI got to the title but tthen crashed
23:31:38[Franklin]the dissolving title?
23:31:55foolshonly there was no disolving
23:32:09[Franklin]ahh... useless animatinos!
23:32:25[Franklin]well, comment out do_title() and see if it works
23:34:08 Quit ZincAlloy (Quit: Leaving.)
23:37:27foolshcongratulations geverts you get a gold star
23:37:37foolshIt works [Franklin]
23:37:43 Quit krabador (Quit: Sto andando via)
23:37:44[Franklin]foolsh: YES!!!!!!
23:38:02*[Franklin] hugs gevaerts
23:38:18[Franklin]finally, good games for rockbox!!!
23:38:31[Franklin]foolsh: how does it work?
23:38:36[Franklin]it shows the protection screen?
23:38:48foolshyes, inputing the code now
23:39:00foolshfeels a bit sluggish
23:39:59 Join krabador [0] (
23:40:00 Quit krabador (Changing host)
23:40:00 Join krabador [0] (~krabador@unaffiliated/krabador)
23:40:04[Franklin]foolsh: who cares?
23:40:05[Franklin]it works!
23:40:22foolshvery cool its running the intro
23:40:35*[Franklin] wants his ipod back
23:40:57*[Franklin] thinks he misplaced it somewhere
23:41:21foolshneeds the screen blanked after visiting the menu
23:41:38[Franklin]yeah, it does
23:41:43[Franklin]who cares?
23:41:44[Franklin]it works!
23:41:47 Join ZincAlloy [0] (
23:41:59foolshaudio clicks
23:42:13foolshseems to run at full speed
23:43:18foolshyes the soundfx are very faint and there is click a prominent clicking
23:43:27[Franklin]foolsh: there are no soundfx
23:43:40foolshthen it just clicks then :)
23:43:43[Franklin]it's a bug
23:43:56[Franklin]well, don't use earbuds
23:44:01[Franklin]just play!
23:44:11[Franklin](and upload a youtube video while you're at it!)
23:44:23foolshlol ok
23:44:38*[Franklin] wants to see this
23:44:50[Franklin]oh yeah, don't forget the credits
23:45:01[Franklin]we need to add all the authors
23:46:22[Franklin]Should I put "Eric Chahi" or "The Another World team" or "The Out of This World team"?
23:47:53[Franklin]there's 7 contributors to xworld, so I'll put "The XWorld team"
23:48:17[Franklin]not sure about newraw
23:50:39[Franklin]I'll put "The Another World team" because the music was composed separately
23:51:20gevaertsThe music is part of the data files I'd expect, so it's debatable whether that belongs in our credits
23:52:10gevaertsI'd go for the team
23:52:48[Franklin]ok then
23:53:03[Franklin]how about newRaw?
23:53:08[Franklin]team or author?
23:55:03*gevaerts doesn't know

Previous day | Next day