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 2012-05-31

00:04:36bluebrotheryay, it's working :)
00:04:53bluebrotherthis byteswapping was swapping too much :o
00:06:10 Quit domonoky (Read error: Connection reset by peer)
00:13:39 Quit funman (Ping timeout: 248 seconds)
00:15:00 Join funman [0] (~fun@rockbox/developer/funman)
00:15:13 Join perrikwp_ [0] (
00:17:56 Quit perrikwp (Ping timeout: 245 seconds)
00:17:56 Quit jfc (Ping timeout: 245 seconds)
00:19:03 Join jfc [0] (
00:22:02 Join Scromple [0] (~Simon@
00:26:40 Quit Strife89 (Quit: Heading home.)
00:29:54bluebrotherdoes voice work in the clip sim? Or doesn't voice work for lowmem targets in the sim?
00:35:14 Join HeartHacker [0] (~social@
00:35:29 Part HeartHacker
00:37:14 Join perrikwp [0] (
00:38:14 Join saratoga [0] (98032909@gateway/web/freenode/ip.
00:39:32 Quit perrikwp_ (Ping timeout: 250 seconds)
00:44:20 Join pamaury [0] (
00:44:21 Quit pamaury (Changing host)
00:44:21 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
00:44:31 Join leavittx [0] (~leavittx@
00:44:58 Quit bertrik (Quit: And That, My Liege, Is How We Know the Earth to Be Banana Shaped)
00:48:53CIA-79Commit ba8e436 in rockbox by Michael Sevakis: Disable attempts to voice init_tagcache.
00:49:29 Quit perrikwp (Read error: Connection reset by peer)
00:50:44 Join perrikwp [0] (
00:51:20 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
00:51:23CIA-79ba8e436 build result: All green
00:51:34 Quit bertrik (Client Quit)
01:05:57 Quit liar (Ping timeout: 245 seconds)
01:08:01 Quit MethoS- (Quit: Konversation terminated!)
01:13:15 Quit AlexP (Read error: Connection reset by peer)
01:16:16 Join AlexP [0] (~alex@rockbox/staff/AlexP)
01:22:28leavittxPeople, I'm trying to run rockbox sim in gdb, but the process receive SIGUSR1 signal. Who knows how to fix that?
01:24:04leavittxJdGordon? :)
01:32:02 Quit remlap (Ping timeout: 244 seconds)
01:33:24leavittxNevermind, I figured it out myself
01:38:01 Quit hillshum (Ping timeout: 256 seconds)
01:45:24jhMikeSsaratoga: oh, there was a gerrit patch lol
01:45:44saratogayeah was wondering about that
01:48:47***Saving seen data "./dancer.seen"
01:51:21 Quit Rower85 (Read error: Connection reset by peer)
01:54:10 Quit perrikwp (Read error: Connection reset by peer)
01:55:23 Join perrikwp [0] (
01:57:05 Join perrikwp_ [0] (
01:59:47 Quit perrikwp (Ping timeout: 252 seconds)
02:02:06 Quit prof_wolfff (Ping timeout: 246 seconds)
02:03:09 Quit AlexP (Ping timeout: 246 seconds)
02:04:36 Join enthdegree [0] (~enthdegre@wikimedia/enthdegree)
02:09:05 Join perrikwp [0] (
02:11:24 Quit perrikwp_ (Ping timeout: 250 seconds)
02:19:23 Quit leavittx (Ping timeout: 256 seconds)
02:25:18 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
02:25:41 Join bitcraft_ [0] (
02:27:26 Quit bitcraft (Read error: Connection reset by peer)
02:31:20 Join hillshum [0] (
02:36:48 Join fyre^OS [0] (
02:38:36 Quit fyrestorm (Ping timeout: 240 seconds)
02:39:59 Join AlexP [0] (~alex@rockbox/staff/AlexP)
02:41:04 Join perrikwp_ [0] (
02:41:54 Quit perrikwp (Read error: Operation timed out)
02:48:59 Join davharster [0] (
02:49:02 Quit pamaury (Remote host closed the connection)
02:51:17 Quit davharster (Client Quit)
03:11:04 Join perrikwp [0] (
03:14:19 Quit perrikwp_ (Ping timeout: 265 seconds)
03:20:05 Join perrikwp_ [0] (
03:22:36 Quit perrikwp (Ping timeout: 240 seconds)
03:25:21 Join B4gder [241] (~daniel@rockbox/developer/bagder)
03:27:46 Quit bitcraft_ (Remote host closed the connection)
03:34:43 Join bitcraft [0] (
03:48:51***Saving seen data "./dancer.seen"
04:01:49 Quit bitcraft (Remote host closed the connection)
04:02:25 Join TheSphinX_ [0] (
04:05:41 Quit TheSphinX^ (Ping timeout: 248 seconds)
04:06:33 Join bitcraft [0] (
04:12:23 Quit perrikwp_ (Read error: Connection reset by peer)
04:13:35 Join perrikwp [0] (
04:24:21 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:24:21 Quit amiconn (Disconnected by services)
04:24:22 Quit pixelma (Disconnected by services)
04:24:24 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:24:26 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:24:43 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:33:28 Quit bitcraft (Remote host closed the connection)
04:35:38 Quit [7] (Disconnected by services)
04:35:44 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
04:43:31 Join bitcraft [0] (~bitcraft@
05:35:20 Quit anewuser (Read error: Connection reset by peer)
05:37:08 Join B4gd3r [0] (
05:38:08 Quit B4gder (Disconnected by services)
05:38:20 Nick B4gd3r is now known as B4gder (
05:38:24 Join B4gd3r [0] (
05:38:44 Quit B4gder (Changing host)
05:38:45 Join B4gder [241] (~daniel@rockbox/developer/bagder)
05:39:27 Quit B4gd3r (Remote host closed the connection)
05:48:53***Saving seen data "./dancer.seen"
05:49:24 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
05:52:54 Quit [Saint] (Ping timeout: 260 seconds)
06:02:43 Quit nosa-j (Excess Flood)
06:03:14 Join nosa-j [0] (
06:06:17 Quit wtachi (Ping timeout: 245 seconds)
06:16:58 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
06:22:24 Quit perrikwp (Read error: Connection reset by peer)
06:23:37 Join perrikwp [0] (
06:29:25 Join chkktri_ [0] (chikakitaa@unaffiliated/chkktri)
06:30:48 Quit chkktri (Read error: Connection reset by peer)
06:37:27 Join tails___ [0] (
06:38:29 Quit enthdegree (Quit: HydraIRC -> <- s0 d4Mn l33t |t'z 5c4rY!)
06:41:00 Quit chkktri_ (Ping timeout: 248 seconds)
07:03:47 Quit scorche` (Read error: Connection reset by peer)
07:10:40 Join Ward [0] (
07:11:06 Nick Ward is now known as Guest51184 (
07:13:53 Quit Guinness` (Ping timeout: 245 seconds)
07:18:27 Quit Guest51184 (Read error: Connection reset by peer)
07:19:00 Join scorche [0] (~scorche@rockbox/administrator/scorche)
07:48:55***Saving seen data "./dancer.seen"
08:05:32 Join LinusN [0] (
08:05:47 Quit aevin (Ping timeout: 246 seconds)
08:06:14 Join aevin [0] (eivindsy@unaffiliated/aevin)
08:06:29 Quit sc (Ping timeout: 246 seconds)
08:06:34 Join wtachi [0] (
08:06:43 Join sc [0] (
08:08:55 Quit mshathlonxp (Ping timeout: 252 seconds)
08:32:42 Join Zagor [0] (
08:32:42 Quit Zagor (Changing host)
08:32:42 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:35:05 Join wodz [0] (
08:35:50wodz <- this looks pretty simple to use in rb. MPLv2 is GPL compatible I think
08:39:30wodzyeah, says MPLv2 is GPL compatible if not explicitly restricted
08:45:23 Join ender` [0] (
08:49:01 Join n1s [0] (
08:49:01 Quit n1s (Changing host)
08:49:01 Join n1s [0] (~n1s@rockbox/developer/n1s)
08:50:17 Join ks [0] (
08:55:34 Join leavittx [0] (~leavittx@
08:58:10 Quit BHSPitMonkey (Remote host closed the connection)
08:59:52 Nick Totalled is now known as Totalled_ZzZz (
09:04:25kugelwodz: looks cool. it uses malloc but that shouild be fixable
09:05:46wodzkugel: we need place for loading bin anyway
09:06:38kugelbtw, there's firmware/load_code.c but I don't suppose it can be unified with g#190?
09:06:45kugel g#190
09:06:46fs-bluebotGerrit review #190 at,190 : Move load_firmware() to separate file to share it between bootloaders and rolo(). by Marcin Bukat (changes/90/190/1)
09:07:48kugelload_code.c is used for codecs, plugins and imageview.rock
09:09:59wodzheh, technically speaking the loading code is not correct. It assumes that sections are layed out one after another which can be untrue. Should be easy to fix though
09:10:11wodzI mean bflt from the link
09:11:58kugelwodz: iirc it's true for bflt (there's an article about it)
09:12:21kugelin the file anyway
09:12:47wodzkugel: thats not true. .text don't have to be just after bflt header
09:14:43wodzehh no. The problem is that bflt can have .text segment in ROM and .data in RAM
09:15:09wodz"As every absolute address is referenced to zero, we can simply add the base address of the text segment to each address needing fix-up. This would be true if the data segment immediately follows the text segment, but we now have complications of -msep-data where the text segment can be in ROM and the data segment in another location in RAM. Therefore we can no longer assume that the .data+.bss segment and text segment will immediately follow
09:15:09wodzeach other. "
09:16:31kugeli misremembered then
09:21:34 Quit wodz (Quit: Leaving)
09:28:02 Quit Scromple (Quit: Leaving)
09:35:02 Join thegeek_ [0] (
09:35:47 Join chkktri_ [0] (chikakitaa@unaffiliated/chkktri)
09:36:01 Join [Saint__] [0] (~Saint]@
09:36:02 Quit [Saint__] (Changing host)
09:36:02 Join [Saint__] [0] (~Saint]@unaffiliated/saint/x-8516940)
09:36:33 Quit tails___ (Write error: Connection reset by peer)
09:38:04[Saint__] 240x320 CabbieV2 on github should "just work" with RTL langs now, apart from some ugly hardcoding of English strings for a few non-translated things.
09:38:22[Saint__]I really don't know what to do about that.
09:38:23 Quit ThomasAH (Ping timeout: 252 seconds)
09:38:29 Join ThomasAH [0] (
09:38:45 Quit [Saint_] (Ping timeout: 252 seconds)
09:38:45 Quit thegeek (Ping timeout: 252 seconds)
09:48:59***Saving seen data "./dancer.seen"
10:05:34 Join einhirn [0] (
10:27:58 Join remlap [0] (
10:37:46 Quit n17ikh (Ping timeout: 244 seconds)
10:38:53 Join n17ikh [0] (
10:40:32 Join Misanthropos [0] (
10:43:01 Nick [Saint__] is now known as [Saint] (~Saint]@unaffiliated/saint/x-8516940)
10:56:18 Quit B4gder (Quit: Leaving)
11:01:46JdGordonanyone care to comment on g#220?
11:01:59JdGordon g#220
11:02:01fs-bluebotGerrit review #220 at,220 : make the sleep timer menu item usable in quick screen by Richard Fröhning (changes/20/220/3)
11:04:47 Quit Sleepy_Coder (Read error: Connection reset by peer)
11:05:01 Join Sleepy_Coder [0] (~z_Z_z_Z_z@unaffiliated/sleepycoder/x-938672)
11:10:13 Quit georgehank (Ping timeout: 260 seconds)
11:11:37 Join georgehank [0] (
11:14:12 Nick Misanthropos is now known as RFroehning (
11:22:49JdGordonis Richard Fröhning here?
11:23:05*JdGordon looks at the previous nick change and facepalms
11:23:08JdGordonRFroehning: ping?
11:29:00RFroehningJdGordon, hi
11:31:49RFroehningthe second patchset can be deleted, was a mistake i made...
11:32:17 Join wodz [0] (
11:32:43RFroehningone thing about the Start Sleep Timer... there is another Setting "Start Sleep Timer On Boot"... thats why I used Run instead...
11:33:37JdGordonRFroehning: seen my comment?
11:33:39wodzwith a bit of hackery it is possible to install elf2flt and friends and it seems to not break our toolchains
11:37:23RFroehningyes - it true there is no indication about the sleeptimer running or not
11:38:09JdGordonI'm not sure my suggestion will completly fix it, but i tihnk its better than the yes/no thats there now
11:39:06RFroehninginstead of the status bar I used a modified wps which displays the sleep timer when running
11:39:19RFroehningbut an indicator in the statusbar would be nice indeed
11:39:31JdGordonand comepltly seperate issue :)
11:39:38JdGordonah, there is already a skin tag for it?
11:41:49RFroehningI like the cusom formatter, but it has this drawback - it cant go to the quickscreen
11:42:14JdGordonyes it can
11:42:25JdGordoncustom int
11:42:35RFroehningoh - i have to look into it then
11:43:49JdGordonyou'll need use the F_TEMPVAR flag in the settings_list item
11:49:01***Saving seen data "./dancer.seen"
11:54:10JdGordonif you dont then changing from "continue" to on/off will happen immediatly which will probably break things
11:57:39RFroehningok - i need to see how that works - and try to apply your suggestions
12:05:03 Quit ender` (Ping timeout: 265 seconds)
12:05:32 Quit ender| (Ping timeout: 260 seconds)
12:06:41gevaertswodz: pong
12:06:42RFroehningJdGordon, can it be you mean the choice setting instead of custom int?
12:08:10JdGordonwell, either shold do it
12:08:17JdGordonINT_SETTING_W_CFGVALS() is probably best
12:08:49JdGordonyeah, definitly that one
12:09:47wodzwhat am I missing that I get this when hand compiling plugin?
12:10:55wodzgevaerts: I was about to ask about some makefile support but now I dived into something else
12:11:40wodzgevaerts: still maybe you know what I am missing when linking plugin? pluginlib or something
12:12:02gevaertswodz: no idea. Have you compared what you're doing to what make V=1 does?
12:13:03wodzgevaerts: I took the line make V=1 gives and added -Wl,-elf2flt to get bflt format output
12:13:20wodzgevaerts: I guess it adds something at linking stage
12:13:44*gevaerts has a quick look
12:14:15wodzmissing rb symbol pretty much looks like something generic
12:15:25wodzmy hacky idea is to write plugin which will be able to load and run another plugin in bflt format :-)
12:15:47gevaertsThat sounds like a good start :)
12:19:26gevaertswodz: for a gigabeat F build, if I take the line used to build lamp.elf and add -Wl,-elf2flt at the end, it seems to link for me
12:19:46gevaertsI do get "arm-elf-eabi/bin/ld: warning: cannot find entry symbol lf2flt; defaulting to 0000000001f56818", but that's it
12:22:25 Join pamaury [0] (
12:22:26 Quit pamaury (Changing host)
12:22:26 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
12:24:04*Zagor sees 43 hotmail addresses unsubscribed due to bouncing. they are blocking us again, for unknown reasons.
12:24:44gevaertswodz: I wasn't trying to compile straight from the C file, I just replaced (and changed) the line that links everything to make the elf file
12:25:14gevaertsIt looks like you're missing plugin_crt0.o
12:25:39gevaerts(at least)
12:25:46gevaertsanyway, lunch!
12:27:28wodzgevaerts: with plugin_crt0.o I get /home/wodz/crosstoolchains/arm-elf-eabi/bin/ld.real: error: no memory region specified for loadable section `.header' which makes more sense
12:27:46 Quit mystica555 (Remote host closed the connection)
12:28:18 Quit nosa-j (Excess Flood)
12:28:32 Join T44 [0] (
12:28:40 Quit Topy44 (Read error: Connection reset by peer)
12:28:50 Join mystica555 [0] (
12:28:59 Join nosa-j [0] (
12:31:58kugelJdGordon: sleep timer it already uses a custom formatter right now doesnt it?
12:32:04wodzbah, it is way more complicated that I thought at the begining.
12:32:14kugelthe menu item shows the remaining time if it's running
12:34:11JdGordonyes, different formatter
12:34:46JdGordonthere is the formatter for the menu entry (which it uses now, being as it isnt a setting) and the formatter the setting screen uses to display the items
13:02:15 Quit kadoban (Ping timeout: 252 seconds)
13:12:43 Quit leavittx (Ping timeout: 265 seconds)
13:13:03 Quit jfc (Ping timeout: 245 seconds)
13:13:46gevaertswodz: why do you want to compile the entire plugin in one go? Just replacing (or adjusting) the final link line sounds much easier...
13:14:45 Join leavittx [0] (~leavittx@
13:15:26wodzgevaerts: what you mean by final link line?
13:15:36*gevaerts will paste a diff in a minute
13:15:43 Join jfc [0] (
13:15:47 Join Guinness [0] (
13:17:05wodzanyway this will not work I guess. plugin_ctr0.c defines sections which are not known to elf2flt.
13:17:40wodzI'll start with something more basic.
13:18:59gevaertsNot sure if those flt files actually are correct though
13:19:10kugelwodz: yea, you need to handle iram
13:19:47gevaertsYou don't need iram for all plugins, it probably makes sense to ignore that for now
13:19:59kugelthat's why it's a tough gsoc proposal. existing solutions such as blft are not entirely sufficient
13:20:27kugelgevaerts: that means two loaders doesnt it?
13:20:55wodzextending bflt should be easy it is just a matter of making it understand i* sections
13:20:58pamauryI think we should handle device serial numbers in a more proper way. The current way (everything in usb_core.c) is a bit ugly
13:21:00kugelwhat's the cost of relocatable binaries (not PIC)?
13:21:22wodzkugel: startup time I guess
13:21:44Torneimplementation complexity. the startup time will be negligible.
13:21:58gevaertspamaury: go for it :)
13:22:12kugelbut no runtime cost (ignoring code size of the loader)?
13:22:17Tornenone at all
13:22:28Tornerelocations are applied once and then discarded
13:22:40kugelthen we ideally want a single loader for core, plugins and codecs :)
13:22:55Torneyeah, you don't need to implement the loader logic more than once.
13:23:10Tornealso, the core doesnt really need ot be relocatable, still
13:23:13Torneat least, on most platforms.
13:23:47kugelbut having a single loader (and not 2+) is a good thing isnt it?
13:23:49wodzI feel bflt or something similar is the way to go. Our current approach is very limited
13:25:04 Quit jfc (Read error: Operation timed out)
13:25:53gevaertsIf we have a decent loader in core, we could consider splitting the core for (very) lowmem players so you can e.g. get rid of the database code if you're not using it
13:25:54pamaurygevaerts: how do you think we should do that ? I was thinking about a dev_serial.h header and a dev_serial/ subdirectory in drivers/ containing one driver per way of getting it. And of course a fallback one. Currently I think there would be: as3514, imx233ocotp, ipodarch and ata. Is that reasonable? Do you have suggestions?
13:26:31kugelgevaerts: more stuff as plugin generally, yes
13:26:51 Join jfc [0] (
13:27:07Tornekugel: you are misunderstanding what "loader" means in this context, i think
13:27:15Tornethe thing that actually does the loading a binary into ram is not interesting
13:27:18Tornethat's just read()
13:27:27Tornethe relocation part is the intersting bit
13:27:33Torneand that part just doesn't need to be done for core
13:27:38Tornethough we could, indeed, split core into more parts
13:27:47Tornethe "main" binary shouldn't ever need to be relocatable
13:28:15kugelyes, but if it can be at no cost we don't need to maintain multiple loaders
13:28:22gevaertspamaury: sounds reasonable to me, although I'd call it something else ("serial" implies a serial bus to me, usually)
13:28:31Tornei don't think i understand what "multiple loaders" you're talking about
13:28:35kugelit's not just read(), it's crt startup too
13:28:43Torneer, what?
13:28:58Tornei don't see how that's related
13:29:26pamaurygevaerts: hmm, right. Any idea? serial_number? dev_id? serial_id?
13:29:41kugelwe could use the same loader for rolo that loades plugins. currently they are separate (and another one in boootloaders)
13:29:45gevaertspamaury: serial_id sounds best I think. "number" is so long... :)
13:30:03Tornekugel: i don't understand what code there is in there to even combine, though
13:30:19Torneit loads a file from disk to a preset address
13:30:23Tornethen jumps to its entry point
13:30:26Tornethere isn't anything else
13:30:28pamaurygevaerts: does the iSerial usb string needs to be 0 terminated?
13:31:37gevaertspamaury: no. It needs to know how long it is, but that's it. I'd consider having the serial_id api just return a char[] and a lenght, and let the calling code deal with how it wants to format the thing
13:32:17pamaurygevaerts: I'm asking this because the code in usb_core.c seems to set the length of the descriptor to 2+2*(nr_chars+1) each time
13:33:47gevaertspamaury: the length is the length of the entire descriptor, including the descriptor type and length field
13:34:25pamauryyes I know, but have a look at the ramdisk case for example: bLength=84. That's 2+2*41 and the string is only 40 characters
13:35:05pamaurysorry, ata case
13:35:17pamaurybut all cases are doing this
13:37:04pamaury68 => 33 characters, 52 => 25 characters
13:37:09gevaertsAh, right!
13:37:25gevaertsThe first character is for encoding the interfaces we have enabled
13:38:00gevaertsSo we have to allocate one more, and start at wString[1]. wString[0] depends on MSD, HID, serial, ...
13:39:03pamauryah!! There should be a comment about that
13:40:17wodzgevaerts: with your patch plain elf is generated (with bflt extension though)
13:40:35gevaertswodz: yes, something is wrong. It also doesn't give me this warning anymore
13:41:16wodzand if I recall correctly gigabeatF doesn't use iram which would explain lack of errors from this side
13:43:06gevaertsoh, of course
13:43:24wodzgevaerts: its −−Wl,-elf2flt versus -Wl,-elf2flt :-)
13:43:40gevaertsYes, I just saw :)
13:46:37wodzok now it throws error from elf2flt at least
13:49:02***Saving seen data "./dancer.seen"
13:57:35CIA-79Commit a87a9ef in rockbox by Amaury Pouly: imxtools: move tools to a new sbtools/ subdirectory
13:57:36CIA-79Commit 64b8d1e in rockbox by Amaury Pouly: imx233: add ocotp driver
13:57:37CIA-79Commit a966763 in rockbox by Amaury Pouly: imx233: add ocotp debug screen
13:57:37DBUGEnqueued KICK CIA-79
13:57:37CIA-79Commit 9bae382 in rockbox by Amaury Pouly: Fill USB serial number descriptor on imx233 targets.
13:57:59pamaurygevaerts: I will implement the serial_id framework and post in on gerrit for review later
13:59:58CIA-799bae382 build result: All green
14:00:18gevaertspamaury: sounds good
14:04:22 Join dfkt [0] (dfkt@unaffiliated/dfkt)
14:12:27pamauryhmm, that's incredible how the mp3 -> flac transition on the fuze+ *always* causes a data abort or a prefetch abort
14:39:38 Join lebellium_gs2 [0] (
14:41:40wodzwhere is rb ptr defined which all plugins refer to?
14:43:21wodznevermind its in apps/plugin.c
14:44:25lebellium_gs2In sbs what are the possible entries for "show icons:" ? Cause when I load the Alien+ theme on R0 then reload my theme the icons are not shown anymore
14:53:04pamauryhmm, I can't a ogg file anymore on my device, the output is just garbage
14:56:25 Join TheLemonMan [0] (
15:01:18lebellium_gs2Euh noy sbs, cfg file I mean
15:01:49wodzplugin.c which defines const struct plugin_api *rb is compiled into core. How does linker do not barf about missing symbol when linking plugins then?
15:05:42Tornearen't plugins linked with -r?
15:08:39 Join XavierGr [0] (
15:08:45 Quit XavierGr (Changing host)
15:08:45 Join XavierGr [0] (~xavier@rockbox/staff/XavierGr)
15:08:57wodzLDFLAGS are like this -T/home/wodz/rockbox-git/build/apps/plugins/ -Wl,−−gc-sections -Wl,-Map,apps/plugins/
15:10:27wodzwell PLUGINLDFLAGS
15:12:31Tornewodz: you're mistaken, is why :@)
15:12:44Tornethere *is* a const struct plugin_api *rb in core
15:12:54Tornewhich is a pointer to the one in core, i think
15:13:07Tornebut there is also one in each plugin, defined in PLUGIN_HEADER
15:13:24Torneand then further pointed to by struct plugin_header for plugin loader to find it
15:14:03Tornehm, actually, no, there's not a "rb" in core that i can see
15:14:17TorneNot there
15:14:20Torneat least, not in the copy i'm looking in
15:14:30Tornethere is an actual instance of the struct called rockbox_api
15:14:51wodzhmm true
15:15:22Torneso yeah. *rb is defined in each plugin by PLUGIN_HEADER, and is defined as extern in plugin.h
15:15:36Torneand then the .header section of the plugin points to it, so that core can assign rockbox_api's address to it
15:16:49wodzI apparently miss the step where *rb is set to address in the core
15:16:59wodzor don't understand it to be precise
15:16:59Torneit's done via an additional pointer
15:17:14wodzyou mean by the loader?
15:17:24Tornethe plugin header has a field "api"
15:17:28Tornewhich is a pointer to *rb
15:17:57Torneplugin_load just does *(p_hdr->api) = &rockbox_api;
15:18:01 Quit lebellium_gs2 (Quit: Bye)
15:18:10Tornethe header is compiled into a separate section called .header
15:18:16Tornewhich our linker script alwasys places at the beginning
15:18:18Torneso the loader can find it
15:19:05Torne"*(p_hdr->api)" from core's POV is "rb" from the plugin's POV
15:19:22wodzok I think I got it
15:19:32Tornethis is, basically,a relocation :0
15:19:38Tornejust a very limited one that can only do one thing :p
15:23:06wodzok our toolchain produce R_ARM_V4BX realocs which elf2flt doesn't handle
15:26:55pamaurywhat is the linker script used by the codecs?
15:27:59Tornewodz: elf2flt should just discard those relocations unless the target is ARMv4 (not ARMv4T)
15:28:07TorneI dont' think we have any ARMv4 targets.
15:28:52Tornethey are used to mark where bx instructions occur, so that the linker invoked with −−fix-v4bx can convert them to "mov pc, <rN>"
15:29:00Tornein order to make binaries run on chips without thumb
15:29:40wodzTorne: I am just trying to do this :-)
15:29:56Tornewell, patch elf2flt :p
15:34:48 Quit wodz (Quit: Leaving)
15:47:58 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
15:49:06***Saving seen data "./dancer.seen"
15:50:52 Quit jfc (Ping timeout: 244 seconds)
15:54:58 Part LinusN
15:55:44 Join jfc [0] (
15:56:55CIA-79Commit cc59ea4 in rockbox by Amaury Pouly: Fix tool for codecs
15:58:51CIA-79cc59ea4 build result: All green
16:19:57 Join msh [0] (msh@
16:20:44 Join MethoS- [0] (~clemens@
16:26:16 Nick msh is now known as mshathlonxp (msh@
16:34:42 Quit Guinness (Read error: Connection reset by peer)
16:36:46 Quit hillshum (Ping timeout: 246 seconds)
16:36:48 Join Guinness [0] (
16:40:34 Quit leavittx (Quit: Ухожу я от вас (xchat 2.4.5 или старше))
16:43:11 Quit bitcraft (Read error: Connection reset by peer)
16:43:30 Join bitcraft [0] (~bitcraft@
16:49:19 Join ender` [0] (
16:49:29 Join ender| [0] (
16:49:51 Quit ps-auxw (Disconnected by services)
16:49:59 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
16:57:11 Join Kaiscene_ [0] (
17:00:20 Quit Kaiscene (Ping timeout: 252 seconds)
17:10:16 Quit Elfish (Ping timeout: 272 seconds)
17:10:36 Quit ender` (Read error: Connection reset by peer)
17:11:13 Join ender` [0] (
17:14:12 Join Elfish [0] (
17:18:53 Quit ved (Ping timeout: 260 seconds)
17:19:00 Join ved [0] (
17:20:42 Quit Kohlrabi (Disconnected by services)
17:20:52 Join Kohlio [0] (
17:21:39 Nick Kohlio is now known as Kohlrabi (
17:22:35 Quit user890104 (Ping timeout: 252 seconds)
17:22:35 Quit Zagor (Quit: Clint excited)
17:25:02 Join user890104 [0] (
17:26:48 Join enthdegree [0] (~enthdegre@wikimedia/enthdegree)
17:27:00 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
17:27:26 Join hillshum [0] (
17:29:02 Quit n1s (Quit: Ex-Chat)
17:39:17 Quit bitcraft (Remote host closed the connection)
17:49:09***Saving seen data "./dancer.seen"
17:57:53 Quit ks (Read error: Connection reset by peer)
17:58:27 Join ks [0] (
18:03:09 Quit TheLemonMan (Quit: Does anybody here remembers of Vera Lynn ?)
18:21:13 Join lebellium_gs2 [0] (
18:24:48 Join anewuser [0] (~anewuser@
18:24:48 Quit anewuser (Changing host)
18:24:48 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
18:25:26 Quit factor (Read error: Connection reset by peer)
18:28:07 Quit enthdegree (Read error: Connection reset by peer)
18:29:46 Join enthdegree [0] (~enthdegre@wikimedia/enthdegree)
18:29:57pamauryit is a shame that usb-drv-arc expects that usbbss and qharray have virtual = physical address
18:30:45 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:30:46 Part domonoky
18:31:09 Quit jfc (Ping timeout: 252 seconds)
18:32:14pamauryhaha! The flac -> mp3 data abort on the fuze+ is due to a null-pointer dereference!!
18:32:33pamauryvery weird, prefetch abort at 0
18:34:10 Join jfc [0] (
18:37:02 Quit ps-auxw (Ping timeout: 272 seconds)
18:37:05 Join WalkGood [0] (~4@unaffiliated/walkgood)
18:43:24 Join factor [0] (
18:46:27 Join Wardo [0] (
18:46:36 Join ramides [0] (
18:47:53 Join bertrik [0] (
18:47:54 Quit bertrik (Changing host)
18:47:54 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:56:20 Join lebellium [0] (
18:57:14 Join lebellium_ [0] (
18:58:22 Join lebellium__ [0] (
18:58:30 Quit lebellium_gs2 (Remote host closed the connection)
19:00:37 Quit lebellium (Ping timeout: 246 seconds)
19:00:47 Nick lebellium__ is now known as lebellium (
19:01:14 Quit Wardo (Quit: Blarglarg)
19:01:40 Quit lebellium_ (Ping timeout: 246 seconds)
19:02:15 Join Wardo [0] (
19:03:13bertrikpamaury, I'm fine with committing the RDS improvement patch
19:03:34bertrik g#237
19:03:40pamauryah yeah, I was about to ask you, I forgot about it and stumbled on it on gerrit :)
19:04:18bertriknot quite sure that the current timeout values are completely optimal, but it works fine for me, and better for you :)
19:04:20pamauryI'll commit it right now, feel free to change the magic constants based on experiments if you feel that's it is necessary
19:04:41CIA-79Commit f7b2e31 in rockbox by Amaury Pouly: rds: implement more robust PS/RT parsing
19:05:33 Quit fs-bluebot (Ping timeout: 245 seconds)
19:06:03 Quit bluebrother (Ping timeout: 244 seconds)
19:06:39 Join fs-bluebot [0] (
19:06:59CIA-79f7b2e31 build result: 10 errors, 2 warnings (Amaury Pouly committed)
19:07:47 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
19:08:52bertrikmeh, I don't see how the patch caused that
19:09:08pamauryyeah, that's strange
19:10:22 Quit ks (Quit: Nettalk6 -
19:12:26bertrikI do see the same problem locally, perhaps it's caused by the kernel.h include?
19:13:29pamauryif that is the case, it looks wrong then
19:14:44 Join bitcraft [0] (~bitcraft@
19:16:35pamauryapparently you are right, something is fishy
19:17:21bertrik#include "system.h" fixes it
19:19:16 Quit remlap (Read error: Connection reset by peer)
19:19:28pamauryI'm a bit surprised that it worked before
19:19:51pamauryI'll fix that
19:20:28CIA-79Commit f8aa5c4 in rockbox by Amaury Pouly: Fix red
19:20:29 Join remlap [0] (
19:21:40pamaurybertrik: do you have a non-imx233 device at hand (like the clip zip)?
19:22:23pamaurycould you check (with a recent build) if switching from mp3 to flac or the converse crashes or not?
19:22:28CIA-79f8aa5c4 build result: All green
19:22:44bertriknot sure if I have any flac tracks
19:23:17pamauryyou can easily find one of the internet
19:23:49pamaurybecause of the fuze+, the flac > mp3 switch generates a data abort or an undefined instruction and it's 100% reliable
19:23:51 Join lebellium_ [0] (
19:25:07 Quit lebellium (Ping timeout: 246 seconds)
19:25:17 Nick lebellium_ is now known as lebellium (
19:25:44 Join kevku [0] (
19:26:16 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
19:26:44bertrikok, I'll try
19:26:52bertrikit doesn't crash the sim?
19:28:00pamauryotoh, I can't really see an explaination, except a cache related problem or something broken in the pcm driver
19:30:56bertrikshould I make the flac->mp3 transition naturally, or should I try to skip?
19:31:39bertrikplayback with automatic transition from flac->mp3 seems to work fine
19:31:53pamauryand when skipping ?
19:31:54bertrikI did forward a bit in the flac first though :)
19:32:13bertrikskipping also works fine
19:32:32bertrikI don't have crossfade active, only crossfeed and eq
19:36:52 Quit ramides (Quit: Leaving)
19:39:05 Join TheSphinX^ [0] (
19:39:29 Join GermanMushroom [0] (
19:39:31 Join saratoga_ [0] (98034408@gateway/web/freenode/ip.
19:39:48saratoga_pamaury: you could try disabling the icache and see if that fixes the problem
19:40:48 Join kadoban [0] (
19:41:38 Quit TheSphinX_ (Ping timeout: 272 seconds)
19:42:26 Quit GermanMushroom (Client Quit)
19:42:40 Join GermanMushroom [0] (
19:49:10***Saving seen data "./dancer.seen"
19:53:34 Quit nosa-j (Excess Flood)
19:56:03 Join nosa-j [0] (
20:04:57 Join liar [0] (
20:20:54 Quit shamus (Read error: Connection reset by peer)
20:21:02 Join shamus [0] (
20:26:17 Quit enthdegree (Ping timeout: 256 seconds)
20:31:45 Join TheLemonMan [0] (
20:41:03 Quit bitcraft (Remote host closed the connection)
20:41:14 Join hype [0] (~hype@
20:45:05CIA-79Commit f494c3c in rockbox by Dominik Riebeling: ZipUtil: check zip file external attributes for folder.
20:45:22 Quit benedikt93 (Quit: Bye ;))
20:46:50 Join lebellium_ [0] (
20:47:13CIA-79f494c3c build result: All green
20:49:07 Quit lebellium (Ping timeout: 246 seconds)
20:49:15 Nick lebellium_ is now known as lebellium (
20:50:25 Join prof_wolfff [0] (
21:00:19 Quit mc2739 (Ping timeout: 246 seconds)
21:00:58 Quit [Saint] (Read error: Connection reset by peer)
21:02:11 Quit advcomp2019__ (Read error: Connection reset by peer)
21:02:52 Join advcomp2019__ [0] (
21:02:52 Quit advcomp2019__ (Changing host)
21:02:53 Join advcomp2019__ [0] (~advcomp20@unaffiliated/advcomp2019)
21:31:10 Join enthdegree [0] (~enthdegre@wikimedia/enthdegree)
21:49:11***Saving seen data "./dancer.seen"
21:57:55CIA-79Commit dce80ea in rockbox by Bertrik Sikken: (Author: Peter Lecký) FS #12686 - Slovak lang update
22:00:32 Quit anewuser ()
22:03:53funmanbertrik: using -g seems to be unanimely popular
22:04:22 Quit y4n (Quit: Today is the perfect day for a perfect day.)
22:06:00 Join Zeether [0] (
22:07:03ZeetherI'm having an issue with the log and uploading it...I've used two programs to scrobble the log and none of the tracks in it are showing up in my account
22:08:43 Join anewuser [0] (~anewuser@
22:08:46 Quit anewuser (Changing host)
22:08:46 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
22:10:02 Join enth [0] (~enthdegre@wikimedia/enthdegree)
22:11:31 Quit enthdegree (Ping timeout: 244 seconds)
22:11:32 Nick enth is now known as enthdegree (~enthdegre@wikimedia/enthdegree)
22:11:53 Quit TheLemonMan (Quit: Does anybody here remembers of Vera Lynn ?)
22:11:54ZeetherI just tried Rock Scrobbler and it ignored all 7 of the tracks in the log too
22:13:50 Quit WalkGood ()
22:18:06 Quit GermanMushroom (Read error: Connection reset by peer)
22:22:33 Quit shamus (Read error: Connection reset by peer)
22:26:02 Join shamus [0] (
22:26:13 Quit shamus (Remote host closed the connection)
22:27:29ZeetherThere must be some problem with my playlist or the log file, I have no idea
22:27:54 Quit KiwiCam (Quit: Leaving)
22:28:39gevaertsZeether: did you play those tracks for long enough?
22:29:04ZeetherI listened to most of them for more than 50% of the track, which is enough
22:29:05CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:29:05*gevaerts has no idea how this works or why one would want it, but he seems to remember something like a 30 second minimum
22:29:11ZeetherThey are listed in the log too
22:29:23gevaertsAre times correct?
22:29:24ZeetherBut when I try scrobbling it ignores them entirely
22:29:44ZeetherI've listened to the tracks before and they've scrobbled just fine
22:31:00 Join KiwiCam [0] (~KiwiCAM@
22:32:25 Join Rower85 [0] (
22:32:33 Quit hype (Quit: ["Textual IRC Client:"])
22:38:19 Quit kevku (Ping timeout: 248 seconds)
22:38:21ZeetherI'm going to try a few things to see if I can fix it, might reinstall rockbox or something
22:42:32 Quit RFroehning (Quit: Verlassend)
22:42:38 Join hype [0] (~hype@
22:47:34 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
22:48:30 Join Rower [0] (
22:50:15ZeetherOkay, I refreshed the database, turned the log off and back on and now it *seems* to work
22:51:22 Quit Rower85 (Ping timeout: 260 seconds)
23:13:43 Join shamus [0] (
23:14:18 Quit saratoga_ (Quit: Page closed)
23:23:47 Quit thegeek_ (Read error: Connection reset by peer)
23:31:13 Quit Rower (Read error: Connection reset by peer)
23:49:13***Saving seen data "./dancer.seen"

Previous day | Next day