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 2012-07-25

00:04:35 Quit mikroflops (Ping timeout: 245 seconds)
00:06:34aswhy aren't my themes working? when I change the theme the font stays small always
00:06:36 Join mikroflops [0] (
00:10:03 Quit TheLemonMan (Quit: WeeChat 0.3.8)
00:10:14asfor example optimalinfo theme, the size of letters should be very big, and it's very small, like on default theme
00:10:40gevaertsDo you have the font pack installed?
00:10:44bertrikmaybe you need the extra fonts
00:10:49 Quit Rondom (Ping timeout: 272 seconds)
00:11:11 Quit muszek (Remote host closed the connection)
00:15:41 Join TheLemonMan [0] (
00:15:41 Quit TheLemonMan (Changing host)
00:15:41 Join TheLemonMan [0] (~LemonBoy@unaffiliated/thelemonman)
00:16:56 Join wodz [0] (
00:19:00wodzgevaerts: thanks for testing. 0x709ec is actual memory access in safe_read32() function used by unwarminder which means it crashed earlier and tried to unwind stack which pointed to some faulty address.
00:19:16gevaertsah, interesting
00:20:04gevaertsI did find that same address a bit suspicious :)
00:20:36wodzanyway v10 is a progress still
00:20:53bertrikwodz, v10? what are you working on?
00:21:04wodzbflt plugins
00:21:14 Join Wardo [0] (
00:21:18fs-bluebotGerrit review #286 at : bFLT loader for plugins by Marcin Bukat (changes/86/286/10)
00:25:08wodzit seems to fix plugins which freezed in v8 but not this which data aborted
00:26:00gevaertsYes, I'd say those "fixed" plugins should check return values a bit more...
00:26:25wodzwhat you mean?
00:26:59gevaertsThey should have handled things a bit more gracefully, like dict does
00:27:12gevaertsi.e. check what get_buffer returns
00:27:18wodzah, yes
00:31:20wodzWell before fix plugin_get_buffer() returned whole plugin buff which lead to plugin code overwrite. dunno why dict chocked on allocation stage but it helped a lot to understand the problem.
00:38:50 Quit LambdaCalculus37 (Quit: LambdaCalculus37)
00:39:11 Quit wodz (Quit: Leaving)
00:39:21 Quit ender` (Quit: MCSE = Must Consult Someone Experienced, Minesweeper Champion, Solitaire Expert, Must Call Someone Else or Moron Claiming System Expertise?)
00:44:09 Join LambdaCalculus37 [0] (~LambdaCal@rockbox/staff/LambdaCalculus37)
00:45:19 Quit as ()
00:47:33 Quit kevku (Ping timeout: 272 seconds)
00:47:59 Join KiwiCam [0] (~KiwiCAM@
00:49:15 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
00:53:07 Join Scromple [0] (~Simon@
01:03:14 Quit anewuser ()
01:06:05 Quit Wardo (Read error: Connection reset by peer)
01:10:06 Quit petur (Quit: Leaving)
01:27:47 Quit mgottschlag (Ping timeout: 248 seconds)
01:37:18***Saving seen data "./dancer.seen"
01:39:30 Quit bertrik (Ping timeout: 252 seconds)
01:51:51 Join linuxstb [0] (
03:13:16 Quit XavierGr ()
03:13:27 Quit LambdaCalculus37 (Quit: LambdaCalculus37)
03:18:06 Quit linuxstb (Quit: This computer has gone to sleep)
03:22:00 Quit TheLemonMan (Quit: WeeChat 0.3.8)
03:37:20***Saving seen data "./dancer.seen"
03:56:16 Quit slowdow (Ping timeout: 252 seconds)
03:56:59 Join slowdow [0] (
04:10:24 Quit [7] (Disconnected by services)
04:10:33 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
04:17:50 Quit lebellium (Quit: ChatZilla [Firefox 15.0/20120717110313])
04:19:18 Quit yuriks_ (Ping timeout: 248 seconds)
04:26:46 Join yuriks [0] (~yuriks@
04:39:18 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:39:18 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:39:18 Quit amiconn (Disconnected by services)
04:39:18 Quit pixelma (Disconnected by services)
04:39:20 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:39:22 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:46:50 Quit Rower85 (Read error: Connection reset by peer)
04:52:48 Quit yuriks (Changing host)
04:52:48 Join yuriks [0] (~yuriks@opentyrian/developer/yuriks)
05:32:24 Quit mikroflops (Ping timeout: 250 seconds)
05:34:27 Join mikroflops [0] (
05:37:22***Saving seen data "./dancer.seen"
06:55:28CIA-5Commit c81bb4e in rockbox by Jonathan Gordon: Initial rockbox.desktop file for Linux DE's
06:57:50CIA-5c81bb4e build result: All green
07:00:11JdGordonwe just need some DBUS support, and something like "$rockbox −−next/prev"
07:05:48 Quit Totalled (Read error: No route to host)
07:06:51 Join Totalled [0] (
07:21:44 Quit mikroflops (Ping timeout: 244 seconds)
07:23:32 Join mikroflops [0] (
07:37:25***Saving seen data "./dancer.seen"
07:59:27 Quit mikroflops (Ping timeout: 244 seconds)
08:12:39 Join mikroflops [0] (
08:27:39 Join Zagor [0] (
08:27:39 Quit Zagor (Changing host)
08:27:39 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:36:31 Quit mikroflops (Ping timeout: 252 seconds)
08:40:47 Join zz_TheSphinX^ [0] (
08:48:43 Join einhirn [0] (
08:55:41 Join mikroflops [0] (
08:57:41 Join ender` [0] (
09:09:00 Join pystar89 [0] (
09:11:42 Join pretty_function [0] (~sigBART@
09:28:05 Join mgottschlag [0] (
09:28:05 Quit mgottschlag (Changing host)
09:28:05 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
09:28:48 Join kevku [0] (
09:30:44kugelJdGordon: we now put files to git root?
09:37:27***Saving seen data "./dancer.seen"
09:40:09 Quit mikroflops (Read error: Operation timed out)
09:48:01 Join wodz [0] (
09:51:33wodzTheSeven: ping again
09:51:55 Join mikroflops [0] (
09:52:03 Quit pretty_function (Ping timeout: 252 seconds)
10:03:24TheSevenwodz: pong
10:03:27[Saint]Is it not slightly weird to put rockbox.desktop (last commit) in the root of the checkout?
10:03:49[Saint]Oh...late to the game.
10:04:48funmanJdGordon: if you want to use DBUS, take a look at MPRIS
10:06:40 Join TheLemonMan [0] (~LemonBoy@unaffiliated/thelemonman)
10:14:49CIA-5Commit 229d88c in rockbox by Frank Gevaerts: Move rockbox.desktop to a better place.
10:15:50JdGordonah damn, didnt see the packaging dir
10:15:54JdGordonthanks for moving it gevaerts
10:16:00gevaertsYou're welcome :)
10:16:59CIA-5229d88c build result: All green
10:19:44 Quit Scromple (Read error: Connection reset by peer)
10:24:48 Quit wodz (Ping timeout: 246 seconds)
10:27:57 Quit [Saint] (Remote host closed the connection)
10:29:11JdGordonhmmm... can FS #10575 be used to do something like "rockbox −−skip"?
10:29:12fs-bluebot Simulator remote-control (patches, new)
10:29:49JdGordonthat should just be open the pipe if it doesnt exist, and if it does connect, send the command and quite no?
10:30:23 Join [Saint] [0] (~quassel@unaffiliated/saint/x-8516940)
10:39:45 Join D_D [0] (
10:39:57kugelJdGordon: what do you want to achieve?
10:40:07kugelsupport multimedia buttons?
10:42:17JdGordonindirectly yeah
10:45:16 Quit mikroflops (Read error: Operation timed out)
10:45:42kugeli think i looked at how rhythmbox does it but i didn't understand it
10:52:13 Join mikroflops [0] (
11:12:10 Join wodz [0] (
11:13:06wodzTheSeven: we can't sync apparently
11:15:57TheSevenwodz: this time it seems to have succeeded though :)
11:16:59wodzTheSeven: how to use postmortem stub and what it gives? Pure mem dump?
11:17:12TheSevenwodz: read access to arbitrary memory locations
11:17:17TheSevenso it can be used to dump some hw state as well
11:17:35TheSevenso you have a panicing nano2g?
11:18:07TheSevenlast time I checked the postmortem stub didn't quite work for some reason, possibly because USB hardware state was screwed up too badly when it started
11:18:15wodzon my own request through g286
11:18:16fs-bluebotGerrit review #286 at : bFLT loader for plugins by Marcin Bukat (changes/86/286/10)
11:18:43 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
11:19:02wodzTheSeven: where are PC tools to interact with the stub?
11:20:06TheSevenno idea where it's hiding, but there should be an file somewhere
11:21:01wodzfind . -name gives nothing
11:21:31TheSevenhere it is
11:22:44wodzagh, Why not to commit it to git tree? With a bit of documentation ;-)
11:22:53TheSevenno idea why that didn't happen
11:23:03TheSevenI thought I had dropped it into the tools folder or something
11:23:19TheSevendoes the post mortem stub itself work?
11:23:27TheSeveni.e. do you see it in lsusb?
11:24:03wodzdon't know - first I wanted to gether all tools needed and failed
11:25:14wodzare there other debugging means emcore will give me? gdbstub or something?
11:28:51TheSevenwodz: emcore does offer some more debugging features, but no gdb stub (yet)
11:30:53wodzTheSeven: What is the cause of the lack of gdbstub? Technical difficulties or just nobody implemented one?
11:31:03TheSevenprimarily the latter
11:31:13TheSevencould be that some more infrastructure in the kernel would be necessary
11:31:29 Quit KiwiCam (Ping timeout: 276 seconds)
11:31:40TheSevenI tried to expose USB APIs for most things required by a debugger stub, but no idea if that's sufficient for GDB
11:31:42TheSevenprobably not
11:33:10TheSevenit provides some basic means of debugging though (logf via usb, memory access via usb, thread register state dump, ...)
11:37:29***Saving seen data "./dancer.seen"
11:40:53wodzTheSeven: I am interested in something different - I would like to run rockbox under control of gdb on target.
11:41:12TheSeventhat would probably require a lot of effort
11:42:21wodzyou can rippoff basic arm gdbstub either from gdbserver of qemu and the bit left is communication driver
11:43:28wodzIn easier version this would be through UART, in more comfortable setup through usb-cdc driver or some custom one + some glue part on pc
11:50:55TheSevenhm, the usb part of that might be rather tricky
11:52:24TheSevenwell the usb driver is a couple thousand lines of code
11:52:43TheSevenlet alone the cdc stuff
11:53:58TheSevenand we would probably need to patch up the rockbox startup code rather badly to make it not break the gdb's usb communication
11:54:57wodzwe could use some simple custom interface and pc glue part
12:00:14TheSevenhow does gdb keep control of the hardware while the application is running?
12:00:19TheSevenby means of a communication IRQ?
12:00:45TheSeventhis might mean that all kinds of critical sections in rockbox might screw us
12:01:41wodzYou usually set brakepoint which jumps to your stub
12:02:18wodzit is up to you how this is implemented. You can use hw assisted brakepoints or simply patch in memory
12:02:42 Join Topy [0] (
12:04:08kugelwodz: /home/kugel/.cross-toolchains/bin/../lib/gcc/arm-elf-eabi/4.4.4/../../../../arm-elf-eabi/bin/ld: cannot find W_RODAT
12:04:52wodzkugel: you mean bflt patch right?
12:06:33kugeljust compiled a e200 build
12:06:41 Quit T44 (Ping timeout: 264 seconds)
12:10:00wodzkugel: strange
12:10:09kugelkugel: seems to choke on the LD battery_bench.rock line
12:11:05wodzkugel: have you installed patched version of toolchain?
12:11:05kugelwhat's required for a build again? i literall just compiled the build
12:11:15kugelright, i havent done that
12:11:22wodzso thats it
12:11:28kugelsorry for the noise then
12:12:00 Join linuxstb [0] ( from the patch should do the trick
12:12:57wodzit needs small manual intervention though - follow on screen instructions
12:13:16kugelcan i still build normal builds with it?
12:14:01wodzit should build normally without magic flag for the linker
12:14:47wodzto be 100% safe you can install toolchain to different dir
12:15:35TheSevenwodz: you might be interested in freemyipod's patched elf2flt
12:15:37 Quit [Saint] (Quit: Quit)
12:15:53TheSeventhat gets away with the standard rockbox toolchain linker
12:15:59 Join [Saint] [0] (~quassel@unaffiliated/saint/x-8516940)
12:16:05 Quit [Saint] (Client Quit)
12:17:50 Join [Saint] [0] (~quassel@unaffiliated/saint/x-8516940)
12:18:16wodzTheSeven: what have you patched? Does it understands .icode, .ncdata and .ncbss magic sections?
12:18:46TheSevenno, there is no such thing in emcore, but I'm sure that could be handled as well
12:19:21wodzSo what did you change?
12:19:38wodzpatched missing relocations or what?
12:20:22TheSevenfixed some bugs, made it work with less weird linker flags, implemented support for ucl compressed binaries, and added some more fields/flags to the header
12:21:06 Join swilde [0] (
12:23:01wodzTheSeven: which bugs?
12:23:34wodzI see you added some missing relocations
12:23:55TheSevenhm, I don't remember the details, that's been years ago
12:24:23TheSevenbut I think it had some issues with some kinds of relocations and some linker flags
12:26:31 Join LambdaCalculus37 [0] (~LambdaCal@rockbox/staff/LambdaCalculus37)
12:27:10 Join Rower85 [0] (
12:33:17 Quit LambdaCalculus37 (Quit: LambdaCalculus37)
12:34:16 Join LambdaCalculus37 [0] (~LambdaCal@rockbox/staff/LambdaCalculus37)
12:34:16 Quit LambdaCalculus37 (Client Quit)
12:34:36wodzhmm the relocation fix stuff is interesting. Need to look at it deeper
12:39:33 Join KiwiCam [0] (~KiwiCAM@
12:43:54 Part swilde ("ERC Version 5.3 (IRC client for Emacs)")
12:45:13wodzTheSeven: did you resign from storing relocation info in network order?
12:46:01 Join lebellium [0] (
12:48:32TheSevenprobably, everything I deal with is LE anyway
12:49:33wodzTheSeven: does freemyipod svn have some more advanced web viewer?
12:52:06wodznevermind - found it
13:01:33 Quit mikroflops (Read error: Operation timed out)
13:05:58 Join mikroflops [0] (
13:22:29 Quit D_D (Quit: Leaving)
13:37:33***Saving seen data "./dancer.seen"
14:04:20 Quit mikroflops (Read error: Operation timed out)
14:12:42 Quit factor (Quit: Leaving)
14:19:41 Quit mgottschlag (Read error: Connection reset by peer)
14:53:36 Join mikroflops [0] (
15:06:00 Join pretty_function [0] (~sigBART@
15:07:43 Quit hype (Ping timeout: 260 seconds)
15:11:04 Join hype [0] (~hype@
15:16:16 Join tchan1 [0] (
15:18:06 Join WalkGood [0] (~4@unaffiliated/walkgood)
15:18:40 Quit tchan (Ping timeout: 264 seconds)
15:21:41 Quit hype (Ping timeout: 250 seconds)
15:23:33 Join hype [0] (~hype@
15:28:38 Join lovasoa2 [0] (
15:29:41 Quit lovasoa2 (Client Quit)
15:37:36***Saving seen data "./dancer.seen"
15:41:53bertrikHaving a file config.h in a codec can conflict with the glboal rockbox config.h, depending on include order, right?
15:50:46 Join XavierGr [0] (~xavier@rockbox/staff/XavierGr)
16:05:22bertrikI better not use it then for now
16:09:14 Quit mikroflops (Ping timeout: 244 seconds)
16:12:58 Join mgottschlag [0] (~quassel@2a00:1398:9:fb00:511f:74dd:accb:4132)
16:12:58 Quit mgottschlag (Changing host)
16:12:58 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
16:17:17 Join mikroflops [0] (
16:22:32 Join petur [0] (~petur@rockbox/developer/petur)
16:31:37 Join lebellium_ [0] (
16:33:00 Quit lebellium (Ping timeout: 246 seconds)
16:33:01 Nick lebellium_ is now known as lebellium (
16:34:09 Join pretty_f_ [0] (~sigBART@
16:35:04 Quit pretty_function (Ping timeout: 244 seconds)
16:36:51 Quit hype (Quit: ["Textual IRC Client:"])
16:40:14 Quit mgottschlag (Read error: Connection reset by peer)
16:40:47 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
16:43:52 Join mgottschlag [0] (~quassel@2a00:1398:200:200:2552:6b7c:3362:f983)
16:43:52 Quit mgottschlag (Changing host)
16:43:52 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
16:54:57 Quit [Saint] (Quit: Quit)
16:55:18 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
16:55:47 Quit pretty_f_ (Remote host closed the connection)
17:01:11 Quit [Saint] (Remote host closed the connection)
17:01:48 Join [Saint] [0] (~sinner@unaffiliated/saint/x-8516940)
17:03:55 Quit Zagor (Quit: Clint excited)
17:11:34 Nick tchan1 is now known as tchan (
17:11:42 Quit tchan (Changing host)
17:11:42 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
17:17:57 Quit amee2k (Ping timeout: 244 seconds)
17:19:46 Join amee2k [0] (
17:37:39***Saving seen data "./dancer.seen"
17:48:48 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:49:26 Join Strife89 [0] (~Strife89@
17:52:09 Quit wodz (Quit: Leaving)
18:05:29 Quit mikroflops (Ping timeout: 244 seconds)
18:07:25 Join mikroflops [0] (
18:08:10 Join einhirn [0] (~Miranda@
18:11:04 Quit petur (Remote host closed the connection)
18:13:41TheLemonManis the internal nand read/writeable by rockbox on a clip zip ?
18:18:47bertrikTheLemonMan, the hardware presents it to us as an SD card, we can't access the raw NAND as far as I know
18:19:40TheLemonManbertrik: so its usable/used by rb, right?
18:20:07bertrikyes, we access it through sd-as3525v2.c
18:21:51 Quit mgottschlag (Ping timeout: 246 seconds)
18:22:00lebelliumdid you get your Zip TLM? What display hw version?
18:24:38TheLemonManits on the way home :D
18:27:28TheLemonManive read that the usb connection might be flaky sometimes
18:28:00lebelliumyes... with my theme most of time the USB connection doesn't work so I use the OF USB
18:29:58TheLemonManhuh? how is that related to the theme :P
18:30:34lebelliumdunno. It also occurs with other themes but I have the impression it occurs more often with mine.
18:51:22 Join saratoga [0] (980329b4@gateway/web/freenode/ip.
18:51:40saratogai think we have some random memory corruption in the theme engine or usb
18:51:42saratogaon the zip
18:51:45saratogamaybe other targets
18:56:56 Join stoffel [0] (
18:57:21TheLemonManworst thing to debug
19:01:53WalkGoodlebellium i've never had an issue with usb other than the first few weeks of release, while i normally don't run a theme, i've been running dfkt's for around 2 weeks now and no issues either
19:03:33lebelliumIndeed, with the other themes it seems to work well. I have had USB issue for 10 months with my theme :) Maybe because my theme uses more memory than the others?
19:03:56lebelliumor maybe due to my 32GB microSD?
19:04:05WalkGoodhave you tried using less to see if your theory is right?
19:04:25WalkGoodi'm running a 32gb card as well
19:04:32 Quit mikroflops (Ping timeout: 276 seconds)
19:04:33 Quit tchan (Ping timeout: 246 seconds)
19:05:08WalkGoodhaven't tried yours, if you want i'll run it next for a few weeks
19:05:24lebelliumso if you have no problem with a 32GB card I would say it's related to the memory corruption saratoga is talking about or theme memory usage more globally
19:05:50WalkGoodthe latter sounds logical
19:06:36TheLemonManmeh modern targets all have 32/64 mb of ram
19:06:59TheLemonManyou usually never run out of memory unless you start using hi res png
19:08:32lebelliumthen I don't know why the Zip freezes 9 of 10 times when I connect it to the computer with my theme :)
19:08:59[Saint]because the Zip doesn't fall into that category? ;)
19:09:09[Saint]Its got nowhere near that RAM does it?
19:10:02lebelliumI don't know, I'm not really familiar with the debug menu
19:12:40[Saint]It _shouldn't_ matter, though, because the theme buffer should be able to infinitely steal from the playback buffer iiuc
19:13:32[Saint]a wps that stole all remaining buffer would kinda defeat the purpose though I imagine.
19:15:49lebelliumJust tried again to make sure.... the Zip won't connect to the computer and when I unplug the cable it remains on the USB connection screen. I have to reset everytime
19:17:57 Join Wardo [0] (
19:20:57 Join mikroflops [0] (
19:22:24saratogaZip has 8 MB of RAM, but tis probably not a lack of ram
19:22:34saratogabut rather something getting double allocated or similar
19:22:51 Join tchan [0] (
19:22:51 Quit tchan (Changing host)
19:22:51 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
19:26:14 Join pretty_function [0] (~sigBART@
19:28:24TheLemonManlolwhat ? 8mb ?
19:28:54AlexPquite a few flash players have small ram
19:29:02AlexPthey don't need it to buffer from disk
19:32:22saratogaclipv1 has 2MB of RAM
19:34:17 Quit bluebrother^ (Read error: Connection reset by peer)
19:34:17 Quit fs-bluebot (Read error: Connection reset by peer)
19:34:21amiconnThe old SH1 archoses all have 2MB too. iFP7xx has 1MB (but that port is abandoned)
19:34:24 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
19:37:42***Saving seen data "./dancer.seen"
19:38:42 Join fs-bluebot [0] (
19:41:54 Quit WalkGood (Quit: ♪ ♫ ♪ ♫ ♪ ♫ ♪)
19:43:10 Join petur [0] (~petur@rockbox/developer/petur)
19:47:21TheLemonManAlexP: no buffering at all ?
19:47:56[Saint]TheLemonMan: of course the buffer, but disk access is a lot faster.
19:48:02AlexPRI don't know, but the point is they don't suffer from spinning the disk up so don't need to buffer large amounts
19:51:21TheLemonManits interesting how sandisk made their firmware tho
19:51:54TheLemonManmodular codecs loaded at runtime, iirc rb has/had the same thing, no ?
19:54:14 Join TomColler [0] (
19:54:26 Part TomColler
19:58:20saratogaTheLemonMan: yes, we dynamically load codecs
19:58:45saratogathe codecs themselves are sort of built on the plugin system which existed first
20:01:44 Join mgottschlag [0] (
20:01:44 Quit mgottschlag (Changing host)
20:01:44 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
20:02:10TheLemonManheh thats how it fits into 8mb
20:05:26 Quit Strife89 (Quit: Heading home.)
20:16:30 Quit tchan (Read error: Operation timed out)
20:33:14 Join tchan [0] (
20:33:14 Quit tchan (Changing host)
20:33:14 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
20:34:01 Join wodz [0] (
20:34:12wodzTheSeven: ping
20:35:58 Quit petur (Quit: Leaving)
20:38:28saratogacodecs are very small though
20:38:34saratogatypically on the order of 30-100KB
20:39:27saratogaabout about 10-15KB for lossless
20:42:34wodzTheSeven: postmortem dumper doesn't work with recent build. The device doesn't show up in lsusb
20:47:43 Join hype [0] (~hype@
20:55:41 Join [Saint_] [0] (~sinner@unaffiliated/saint/x-8516940)
20:56:35 Quit dys (Ping timeout: 272 seconds)
20:56:58 Quit [Saint] (Ping timeout: 276 seconds)
20:59:56 Nick [Saint_] is now known as [Saint] (~sinner@unaffiliated/saint/x-8516940)
21:01:13 Quit [Saint] (Quit: Quit)
21:01:32 Join [Saint] [0] (~sinner@unaffiliated/saint/x-8516940)
21:01:39 Join dys [0] (~user@2a01:1e8:e100:8296:21a:4dff:fe4e:273a)
21:07:21 Quit kevku (Ping timeout: 272 seconds)
21:07:43 Join kevku [0] (
21:09:41 Quit stoffel (Remote host closed the connection)
21:22:15bertrikStill messing around with opus, I get semi-random skip, but it doesn't appear to happen in the opus decoder. Perhaps we're doing something stupid in deframing
21:27:32gmaxwellbertrik: which files are you getting it on?
21:27:46gmaxwelle.g. only the test vectors? or on files you've encoded?
21:27:51bertriksome files I encoded myself with opusenc
21:28:00bertrikI should try the test vectors
21:28:41 Quit hype (Ping timeout: 250 seconds)
21:30:32gmaxwellOkay, the reason I ask is that the testvectors have extensive frame size switching... where the files you encode with opusenc won't but they'll tend to have larger ogg pages.
21:31:07 Join hype [0] (~hype@
21:35:08bertriktestvectorXX.bit.opus in right?
21:37:45***Saving seen data "./dancer.seen"
21:40:44 Quit evilnick (Quit: Leaving)
21:44:17 Quit mgottschlag (Ping timeout: 250 seconds)
21:45:20 Join mgottschlag [0] (
21:45:20 Quit mgottschlag (Changing host)
21:45:20 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
21:45:31 Quit pretty_function (Ping timeout: 255 seconds)
21:48:41gmaxwellbertrik: right.
21:56:36 Join prof_wolfff [0] (
21:56:53 Quit saratoga (Ping timeout: 245 seconds)
22:12:00 Quit wodz (Quit: Leaving)
22:13:23 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
22:15:34 Quit FOAD (Quit: I'll be back)
22:20:10 Join FOAD [0] (~foad@unaffiliated/foad)
22:23:15 Quit GeekShadow (Ping timeout: 248 seconds)
22:33:43 Join n1s [0] (
22:33:43 Quit n1s (Changing host)
22:33:43 Join n1s [0] (~n1s@rockbox/developer/n1s)
22:45:31 Quit mikroflops (Ping timeout: 276 seconds)
22:47:06 Quit hype (Quit: ["Textual IRC Client:"])
23:01:07 Join mikroflops [0] (
23:06:33 Quit linuxstb (Ping timeout: 255 seconds)
23:13:44bertrikmeh, I'm struggling and getting stuck deeper and deeper ...
23:13:58bertrikI'm getting a bad checksum in the ogg stream
23:15:13 Quit mikroflops (Ping timeout: 246 seconds)
23:15:50 Join linuxstb [0] (
23:28:51 Join mikroflops [0] (
23:37:48***Saving seen data "./dancer.seen"
23:38:14gmaxwellbertrik: hey, thats good news.
23:38:54gmaxwellI'm going to bet that your ogg layer is broken then... the maximum size of an ogg page is about 65305 bytes. Perhaps it's not allocating enough for this?
23:42:31bertrikI'm now letting the ogg layer do a malloc/realloc
23:45:56 Join GeekShadow [0] (
23:49:11bertrikin hindsight this will be a silly problem, I know :)
23:50:52gmaxwellif you encode with −−max-delay 0 to opusenc, you'll get tiny pages out... might help diagnose it.
23:53:39 Join factor [0] (
23:53:42bertrikaha! our realloc is really a stub for malloc ....
23:55:18 Quit n1s (Quit: Ex-Chat)
23:56:11gmaxwellha. that'll burn memory fast.
23:57:25 Quit kevku (Ping timeout: 248 seconds)

Previous day | Next day