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 2017-05-04

00:01:37user890104http://i.imgur.com/2xQGe8H.png
00:01:49user890104ipod6g simulator running 8086tiny.rock
00:01:55user890104booting freedos :)
00:02:35*user890104 builds it for the target
00:05:42__builtinheh, nice!
00:06:30user890104that's why i'm trying to get the vagrant setup to work
00:06:37user890104i'm developing on windows
00:09:22user890104i'm also going to prepare a patch for buildzip.pl to use relative paths when symlinking instead of absolute
00:09:36user890104so it works inside vagrant, and probably networks FSes
00:11:52__builtinwhat's the user/pass for vagrant?
00:12:14user890104ubuntu/vagrant
00:12:47__builtinworks :)
00:12:53user890104or user ubuntu and the private key at rockbox/tools/vagrant/machines/default/virtualbox/private_key
00:13:07user890104can you help me with this: https://pastebin.com/raw/AE2ZUESP
00:13:40__builtinquick fix is to increase plugin buffer size
00:13:48__builtinlong-term you're going to have to consider making it an overlay plugin
00:14:23user890104where's the plugin buffer size defined?
00:14:53user890104#define RAM_SIZE 0x10FFF0
00:15:04user890104the emulator wants about 1 mb
00:15:18user890104i think i can cut it down
00:15:28 Join superluig164 [0] (~superluig@toroon12-1177740892.sdsl.bell.ca)
00:16:09__builtinfirmware/export/config/ipod6g.h
00:16:24__builtinit should have an obvious name, I forget what it is
00:17:05user890104128k ram is just enough
00:19:36user890104well, it's not
00:21:36 Join MrZeus2 [0] (~MrZeus@2a02:c7f:7066:fb00:f076:499a:d76a:95ab)
00:21:38 Join ZincAlloy1 [0] (~Adium@2a02:8108:8b80:1700:5528:53db:2b09:6497)
00:22:34user890104so what's an overlay plugin?
00:22:42 Join igorsk2 [0] (~igorsk@host-85-201-83-23.dynamic.voo.be)
00:24:16 Quit anonus (Ping timeout: 255 seconds)
00:24:48 Quit St0neHead (Ping timeout: 255 seconds)
00:25:41 Join akaWolf1 [0] (~akaWolf@unaffiliated/akawolf)
00:26:48 Join igitoor_ [0] (igitur@2a00:d880:3:1::c1ca:a648)
00:28:10 Quit ShameOnYou (Quit: Textual IRC Client: www.textualapp.com)
00:28:39 Quit ZincAlloy (Ping timeout: 240 seconds)
00:28:39 Quit akaWolf (Ping timeout: 240 seconds)
00:28:39 Quit shrizza_ (Ping timeout: 240 seconds)
00:28:39 Quit igitoor (Ping timeout: 240 seconds)
00:28:39 Quit MrZeus1 (Ping timeout: 240 seconds)
00:28:39 Quit igorsk (Ping timeout: 240 seconds)
00:28:42 Nick akaWolf1 is now known as akaWolf (~akaWolf@unaffiliated/akawolf)
00:29:28 Join shrizza_ [0] (~shrizza@oki-dc-urasoe01-187.glbb.ne.jp)
00:29:56saratogaSammysHP: try this build (disables voltage scaling) http://web.mit.edu/~mgg6/www/rockbox-clipplus-novoltagescaling.7z
00:30:20 Quit igitoor_ (Changing host)
00:30:20 Join igitoor_ [0] (igitur@unaffiliated/contempt)
00:32:23 Join St0neHead [0] (~stonehead@2a01:7e00:e001:3700:6667::2)
00:32:53 Join anonus [0] (~anonymous@citadel.niflheim.info)
00:40:39superluig164Any of you guys know about/are interested in pursuing RockBox on the Ruizu us212a?
00:47:13 Join jhMikeS [0] (~jethead71@d192-24-173-177.try.wideopenwest.com)
00:48:59__builtinuser890104: looks like it's PLUGIN_BUFFER_SIZE
00:49:00 Quit xorly (Ping timeout: 276 seconds)
00:49:14__builtintry changing it to 1MB or greater
00:50:06__builtinan overlay is just a plugin loaded by another plugin, usually into the audio buffer
01:00
01:00:56 Quit ender` (Quit: drug, n: A substance that, injected into a rat, produces a scientific paper.)
01:01:01__builtinwe really need to do *something* that gives us a working build system
01:01:35__builtincommitting "blind" or minimally tested things isn't sustainable
01:02:46 Quit alexweissman (Remote host closed the connection)
01:05:16 Join alexweissman [0] (~alexweiss@2001:18e8:2:28b6:98d9:f47c:617e:6d9b)
01:21:25***Saving seen data "./dancer.seen"
01:26:26 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
01:35:39 Quit gevaerts (Ping timeout: 240 seconds)
01:36:16 Quit evilnick_ (Ping timeout: 255 seconds)
01:41:58 Quit superluig164 (Ping timeout: 260 seconds)
01:48:39 Join evilnick_ [0] (~evilnick@84.195.65.126)
01:56:28 Quit alexweissman (Remote host closed the connection)
02:00
02:05:01 Join alexweissman [0] (~alexweiss@2001:18e8:2:28b6:30a4:726a:c1ee:e422)
02:05:24saratogagoogling that player brings up people saying its ATJ2127, so probably not hackable
02:09:47 Join superluig164 [0] (~superluig@okvlon0102w-lp130-02-184-147-154-219.dsl.bell.ca)
02:11:46 Join The_Prospector [0] (~The_Prosp@unaffiliated/cornman)
02:12:48 Quit MrZeus2 (Ping timeout: 255 seconds)
02:58:36 Quit alexweissman (Remote host closed the connection)
03:00
03:20:43 Join alexweissman [0] (~alexweiss@2001:18e8:2:28b6:4cc0:ae1d:9929:f169)
03:21:28***Saving seen data "./dancer.seen"
03:22:16 Quit ZincAlloy1 (Quit: Leaving.)
03:30:45 Quit Bilgus (Remote host closed the connection)
03:31:10 Join Bilgus [0] (~Bilgus@gateway/tor-sasl/bilgus)
04:00
04:12:28 Quit alexweissman (Remote host closed the connection)
05:00
05:03:47 Join igorsk [0] (~igorsk@host-85-201-83-23.dynamic.voo.be)
05:06:47 Join SovietShaman_ [0] (quasselcor@97-87-177-85.dhcp.stls.mo.charter.com)
05:10:34 Join sth [0] (~ZNC-SRV-H@116.111.121.211)
05:11:45 Quit igorsk2 (*.net *.split)
05:11:45 Quit amayer_ (*.net *.split)
05:11:45 Quit CommunistWitchDr (*.net *.split)
05:11:46 Quit dys (*.net *.split)
05:11:46 Quit duo8 (*.net *.split)
05:11:46 Quit preglow (*.net *.split)
05:19:23 Join amayer_ [0] (~amayer@mail.weberadvertising.com)
05:21:31***Saving seen data "./dancer.seen"
05:22:56 Join alexweissman [0] (~alexweiss@149-160-162-52.dhcp-bl.indiana.edu)
05:27:35 Quit alexweissman (Ping timeout: 240 seconds)
05:47:35 Quit krabador (Remote host closed the connection)
06:00
06:58:07 Quit TheSeven (Disconnected by services)
06:58:17 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
07:00
07:21:34***Saving seen data "./dancer.seen"
07:23:34 Join qud0 [0] (1f81dff8@gateway/web/freenode/ip.31.129.223.248)
07:32:24 Quit PurlingNayuki (Ping timeout: 258 seconds)
07:32:59 Join PurlingNayuki [0] (~Thunderbi@2001:da8:215:3c02:14ea:f890:267b:9ef9)
07:41:13 Quit superluig164 (Read error: Connection reset by peer)
07:53:50 Join alexweissman [0] (~alexweiss@c-68-51-123-75.hsd1.in.comcast.net)
08:00
08:20:03 Join ender` [0] (krneki@foo.eternallybored.org)
08:24:38 Join parchd [0] (~parchd@unaffiliated/parchd)
08:26:05 Join PurlingNayuki1 [0] (~Thunderbi@114.255.40.42)
08:29:14 Quit PurlingNayuki (Ping timeout: 260 seconds)
08:31:39 Quit PurlingNayuki1 (Read error: Connection reset by peer)
08:32:05 Join PurlingNayuki [0] (~Thunderbi@114.255.40.42)
08:40:50 Join wodz [0] (~wodz@iwl138.internetdsl.tpnet.pl)
08:47:14 Quit neutric (Quit: WeeChat 1.6)
08:56:19 Quit ender` (Quit: 'And you have to shout -' He tried to remember some far-off reading. '- er, bonsai. Yes. Bonsai!' — Terry Pratchett: Reaper Man)
08:59:30 Join petur [0] (~petur@rockbox/developer/petur)
09:00
09:11:58 Join ender` [0] (krneki@84.255.206.8)
09:21:36***Saving seen data "./dancer.seen"
09:23:31jhMikeSwodz: you pinged me the other day. I pinged back but haven't seen you.
09:26:23wodzjhMikeS: Hi, yes. My question is about switch_thread() or load_context() more precisely. load_context() is implemented as inline asm in arch specific file. Now this is inlined in switch_thread(). I don't quite understand the purpose of running branch in load_context().
09:26:53wodzjhMikeS: Is it used to distinguish between started thread and this just about to start or something else?
09:26:59 Quit PurlingNayuki (Ping timeout: 260 seconds)
09:27:34wodzjhMikeS: What puzzles me is that since this load_context() is inlined switch_thread() epilogue overwrites some of restored registers
09:30:32 Join dys [0] (~dys@ip-109-40-3-86.web.vodafone.de)
09:30:43jhMikeSthe check is sort of a nasty thing so it doens't run the startup routine over and over
09:31:31qud0 Tell me where in the config to store the speed parameter of the song ?
09:31:54jhMikeSthe compiler also isn't really aware of what's going on with those code blocks so just pushes and pops stuff like a normal function call
09:32:31 Quit parchd (Quit: reboot)
09:33:27pamauryqud0: not sure what you mean?
09:35:21qud0I have broken the screen on the player, and the music plays very quickly, I want to change this value through the config
09:36:19[Saint]Config values are detailed in the wiki and manual.
09:36:48qud0That I can not find there this value
09:36:58qud0:-(
09:37:21[Saint]I don't believe the speed is stored in the config.
09:37:31[Saint]So...that'd be why I would guess.
09:37:49wodzjhMikeS: ok, so 'running branch' is regular code path, right?
09:38:09[Saint]AFAIR it's stored in a bookmark?
09:43:56 Join xorly [0] (~xorly@wced-228-217-32-147.feld.cvut.cz)
09:44:35jhMikeSwodz: if by "running branch" you mean "when it tests null", it's just a load of what's saved in the context area
09:47:42wodzjhMikeS: Thats clear. This should cause return from switch_thread() to movet to other thread, If I understand the logic correctly. So if return register (LR in arm for example) is overwritten by epilogue how does it work?
09:48:34 Join elensil [0] (~edhelas@535580F9.cm-6-6c.dynamic.ziggo.nl)
09:51:58jhMikeSwodz: the stack is switched too
09:53:49jhMikeSso, whatever is popped by the compiler is from the stack of the thread just picked to run
09:53:59wodzjhMikeS: ok so this relies on the fact that store_context() is called after prologue and hence double restore is correct?
09:55:29jhMikeSwodz: yeah, it's a bit dodgey at this time and it could go wrong
09:56:08wodzjhMikeS: It apparently goes wrong on mips with -fPIC :P
09:56:39jhMikeStry adding a memory clobber to the store/load? it should make a compiler fence
09:58:06wodzjhMikeS: What you mean by memory clobber?
09:58:46jhMikeSlike the clobber list for registers, add "memory"
09:59:40pamaurywouldn't it be simpler to write it entirely in assembly?
09:59:59jhMikeSwhat? the whole switch thread logic?
10:00
10:00:38 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
10:02:10pamauryload_context. Because on MIPS currently it is inline asm inside a function. I don't know how safe this is (the compile could make assumption on prolog/epilog)
10:02:37pamauryI assume that's what wodz is referring to
10:02:37jhMikeSthere was another way I messed around with
10:03:23kugelthere's a gcc attribute to avoid prolog/epilog, so it could be moved to the inline asm for certainty
10:03:25wodzpamaury: This is only half of the problem. I don't quite understand what goes wrong but it has something to do that switch_thread() overwrites some of restored registers
10:03:26jhMikeSpamaury: I think "memory" could help that since, as I understand it, it won't move loads and stores across that
10:03:32wodzkugel: not on mips :-)
10:03:35jhMikeSworth a try anyway
10:04:02kugeloh, that's sad :)
10:04:05wodzkugel: MIPS people see naked attribute as evil and refused to implement that
10:04:18kugelidiots
10:04:58[Saint]...a wild kugel appears...
10:05:13kugelmaybe make the function appear as a leaf?
10:05:26*[Saint] used commit
10:05:33[Saint]...it's not very effective
10:09:35jhMikeSkugel: I had something that did that. I just marked it "noreturn", internally only and did context save/restore external to it
10:10:07pamaurythen I agree with jhMikeS that could try to put some memory barrier before calling store context and load context
10:10:27wodzhmm, or I could explicitly return in load_context() that way I'd avoid switch_thread() epilogue
10:11:27jhMikeSwodz: better return with the non-volatiles as they were before the function was called though.
10:11:59wodzAnyway I find thread switching code a bit hacky
10:12:26kugelnoreturn causes extensive optimizations at the call site too, maybe unwanted
10:12:56pamaurynoreturn sounds like a bad idea if the function actually returns...
10:13:01jhMikeSkugel: that's why you mark it inside thread.c but not in the header
10:13:36kugelmy understanding of the switch_thread was that only callee-saved regs don't need to be saved and it looked pretty safe to me
10:13:37jhMikeSit won't actually return, load_context effectively long jumps
10:14:09jhMikeSand it gets rid of the null check
10:15:34kugelwell it returns from the POV of the thread that is switched to
10:16:21kugelfwiw, it should be possible to implement the thing without asm and just setjmp/longjmp. worth a try?
10:16:36pamaurywell tell me if I'm wrong but this kinds of assume that the prolog of store_context is compatible with the epilog of load_context. Because store_context will do *things* then save all important registers (including RA). Then load_context will restore registers (thus RA) do *thing* then return (thus longjump) no?
10:17:37pamauryso this is not truly setjmp/longjmp in fact
10:17:50kugeliiuc yes, which should be OK, since both are ordinary functions
10:19:10kugelwhat's wrong on mips+fpic? is something not saved that better should be?
10:19:50pamaurybut then, isn't it the case that this requirement also applies to switch_thread? Because contrary to setjmp/longjmp, the store/load points are different, I can totally see being wrong
10:20:27*pamaury is a bit confusing about how everything fits together
10:21:34pamaury*confused
10:22:49pamauryI think it could be way clearer/more obviously correct if it was used like setjmp:
10:22:49pamauryif(store_context())
10:22:49pamaury return; /* returning from load */
10:23:06pamauryelse { /* switch thread logic */ longjmp(target_thread); }
10:24:14wodzon mips there are a few orthogonal problems 1) when fPIC is used gp should be saved as well (it was not as mips targets didn't used fPIC) 2) fPIC convetnition dictates that t9 needs to hold called function address (it was not) 3) there is some problem with double restore which I don't quite understand BUT creating threads works while first call to switch into already started thread crashes
10:25:03kugelbasically, store_context() saves the regs as entered into switch_thread() (including RA, to the caller of switch_thread()). load_context() restores that, including RA, making switch_thread() effectively return to the caller
10:25:45kugeli forgot that load/store_context are inline, so no prolog/epilog
10:27:21 Join PurlingNayuki [0] (~Thunderbi@2001:da8:215:4ff:e1ba:bb8d:2f24:58d7)
10:27:26pamauryit's not clear to me this is safe, the function could be doing some stack/variable changes before store_context() and those are not undone after load_context() because it's directly returning to caller of switch_thread()
10:27:41pamauryby "the function could ..." I mean switch_thread
10:30:13kugelthere is still the epilog of switch_thread
10:31:03kugelso stack allocations should be undone
10:32:17pamaurybut this making an assumption because precisely load and store are not at the same point in the function. For stack is seems safe because the compiler will only modify the stack once per function, but maybe other things are happening and they explain the problem wodz is seeing
10:33:35*kugel talks without looking at the code actually, this from my memories of 6 years ago :o
10:33:40wodzFirst problem is that mips has not naked attr, so store_context() will move stack
10:34:09pamauryI think super hackish because store_context() and load_context() are inlined
10:34:16pamaurythis makes it completely unsafe
10:34:26pamaurywe have *no idea* what the final code looks like
10:34:41kugelwodz: it should be inlined, so no stack modifications?
10:35:14wodzneed to double chack
10:35:20wodz*check
10:36:41wodzI was really surprised that we tolerate double restores while in other parts of code we squash every single op
10:37:00kugelpamaury: you can force inline so, I'm not seeing what it makes unsafe
10:38:19wodzkugel: actually you can't force inline. Gcc treats inline as a hint
10:39:03kugelthere's another attribute for that, or does mips also not implement that?
10:39:54kugelplus, gcc can be counted on in this case, if it's static and there is only one call site (in my experience)
10:41:10kugelwodz: I couldn't follow from the logs, where is the double restore?
10:44:20wodzkugel: after context_load() epilogue of switch_thread() overwrites some of the registers
10:44:36kugelI wonder...it should be possible to replace the call to store_context() with (pseudo code) "if (setjmp(&current->regs) == 0) return;" and the call to "longjmp(&new_thread->regs);"
10:45:25wodzIsn't that what pamaury proposed?
10:45:29pamaurykugel: the current code is unsafe imo because load and store points are not the same. This is contrast with
10:45:29pamauryif(setjmp() == 0) return; else bla();
10:45:45pamaurywhere wether you store or load, you are at the same point (ie after the test)
10:45:58kugelok, so it's not exactly a double restore, just that a function epilog may overwrite selected regs. I think that only non-callee-saved regs are stored for this reason?
10:46:16*pamaury tries to build an unsafe example for MIPs to illustrate the problem
10:46:47igorskmaybe asm does not specify clobbered regs properly?
10:47:26kugelpamaury: "bla()" is the core of switch_thread(), the scheduler, no need to move that a separate function
10:48:52pamauryby bla(); I meant any other code of the function, not necessarily a function ;)
10:49:49kugelright. I omit the else { ... } after if-return by nature :p
10:50:25pamauryyes you are right, I would write it like this too
10:51:05kugelsetjmp/longjmp would probably safe more than necessary but other than that it should work OK without asm
10:53:39kugelin fact, thread-unix.c does pretty much that, except it does setjmp also in load_context() because putting return in the inline function doesn't have the same effect as putting the return in switch_thread(). If it was written as a macro it could be written as above
10:55:01kugelif store_context() in thread-unix.c was written as a macro*
10:55:48pamaurythis does not mean it's fullproof. the code works on ARM native and hosted but fails on MIPS...
10:56:26kugelhosted or native mips?
10:56:56 Quit PurlingNayuki (Ping timeout: 245 seconds)
10:57:37pamauryhosted (at least)
10:57:47pamaurywodz will confirm I think
10:58:06kugelusing thread-unix.c, i.e. setjmp/longjmp?
10:58:59kugelon hosted the OS might impose restrictions (see ART on android)
10:59:52wodzkugel: AFAIK thread-unix.c is used when HAVE_SIGALTSTACK_THREADS is used as threading model. I tried that. It works in main binary but crashes when loading codecs.
11:00
11:00:19wodzthe platform is linux + glibc
11:03:02pamauryalso the code in thread-unix.c does the right thing: it does not store the context in store_context(). It merely remembers *where* to store it and then in load_context() it does the equivalent of if(setjmp(old_ctx)) longjmp(new_ctx);
11:03:02pamauryI don't know why it fails on MIPS...
11:05:04wodzMaybe this is separate problem. Main binary works with SIGALTSTACK and this switches threads quite regularly
11:05:37pamaurymaybe there is a bug in the codec crt0.S code that overlooks the GOT problem?
11:06:20pamauryis the entirely code compiled with -fPIC ?
11:10:00kugelalso, are both compiled with DEBUG (or both without). this is a classic one
11:10:57kugelis crt0.S used on hosted builds? I believe we compile standard shared libraries
11:11:23pamauryI think I added a check/fix about that (because the DEBUG flag was not carried over to codecs/plugins)
11:11:40*pamaury checks he actually pushed that to mainline
11:13:17wodzkugel: lib/rbcodec/codecs/codec_crt0.c is compiled on hosted
11:14:06kugeloh right. but I remember it contains some #ifdefs for hosted/simulator builds
11:15:47wodzkugel: yes it is greately reduced but still entry point is not taken from elf
11:16:03pamaurywodz: I just remember now: codecs are crashing on ARM on hosted build
11:16:19pamauryOn Sony NWZ linux port, I can't seem to make them work with DEBUG
11:16:27 Quit qud0 (Quit: Page closed)
11:17:06wodzpamaury: I thought it was working on NWZ. How did you fiddled with volume then?
11:17:26pamaurywodz: it's only working on non DEBUG
11:17:28pamaurybuilds
11:17:56wodzok, I am running non debug and it crashes on start
11:21:37***Saving seen data "./dancer.seen"
11:23:29wodzpamaury: btw, There is no possibility to mix PIC and non PIC code on mips due to different convention. Linker barfs when you try to mix objects
11:24:28pamauryok, just wanted to make sure the linker was doing the check and you couldn't mix by mistake
11:28:47 Join paulk-elm [0] (~user@147.210.204.186)
11:31:26SammysHPdon't want to interrupt you, but is here always so much activity?
11:35:04pamauryit depends
11:35:27SammysHPand for the last four years I though rockbox is dead… :/
11:36:09SammysHP(but are there any good players with rockbox support that are still available?)
11:45:24 Quit marex-cloud ()
11:45:48 Join marex-cloud [0] (sid137234@gateway/web/irccloud.com/x-yqiumwfqdrfqsukw)
11:46:04 Join Rower [0] (husvagn@d83-183-134-99.cust.tele2.se)
11:48:52 Quit Rower (Client Quit)
12:00
12:01:56 Join robertd1 [0] (~root@201.242.174.75)
12:03:12 Join TheLemonMan [0] (~root@irssi/staff/TheLemonMan)
12:07:48 Join preglow [0] (~thomj@2001:840:4243:3::101)
12:11:38 Quit paulk-elm (Quit: Leaving)
12:19:51jhMikeSkugel, wodz: here's the hastily cobbled-together way I was thinking (just for ARM): https://pastebin.com/WnEMUUGp
12:20:00 Join krabador [0] (~krabador@unaffiliated/krabador)
12:22:42 Quit xorly (Ping timeout: 246 seconds)
12:43:16 Join xorly [0] (~xorly@wced-228-217-32-147.feld.cvut.cz)
13:00
13:11:55 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:25e5:4f4e:f197:cb7d)
13:21:39***Saving seen data "./dancer.seen"
13:24:56 Join PurlingNayuki [0] (~Thunderbi@2001:da8:215:4ff:e1ba:bb8d:2f24:58d7)
13:31:35 Quit PurlingNayuki (Ping timeout: 258 seconds)
13:45:47 Join p3tur [0] (~petur@91.183.48.77)
13:45:48 Quit p3tur (Changing host)
13:45:48 Join p3tur [0] (~petur@rockbox/developer/petur)
13:48:57 Quit petur (Ping timeout: 240 seconds)
13:49:04 Nick p3tur is now known as petur (~petur@rockbox/developer/petur)
14:00
14:06:04SammysHPsaratoga: your build without voltage scaling doesn't work either :/
14:09:34wodzthe more I dig in codec crash on agptek the less I understand :/
14:14:08 Quit [7] (Ping timeout: 258 seconds)
14:14:24 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
14:21:52 Join ShameOnYou [0] (~SOY@2.1.198.126)
14:21:52 Quit ShameOnYou (Changing host)
14:21:52 Join ShameOnYou [0] (~SOY@unaffiliated/shameonyou)
14:30:57 Quit xorly (Ping timeout: 240 seconds)
14:32:32 Join qud0 [0] (~root@31.129.223.248)
14:36:37 Quit qud0 (Quit: Leaving)
14:38:27 Quit pamaury (Ping timeout: 240 seconds)
15:00
15:21:40***Saving seen data "./dancer.seen"
15:30:58 Quit wodz (Ping timeout: 260 seconds)
15:47:50 Join paulk-elm [0] (~user@147.210.204.186)
15:53:39 Quit kugel (Ping timeout: 260 seconds)
15:53:46 Join kugel [0] (~kugel@rockbox/developer/kugel)
16:00
16:00:56 Quit paulk-elm (Remote host closed the connection)
16:09:58 Quit amayer_ (Quit: Leaving)
16:10:19 Join amayer [0] (~amayer@mail.weberadvertising.com)
16:21:27 Quit jhMikeS (Ping timeout: 240 seconds)
16:23:48 Join Strife1989 [0] (~quassel@adsl-98-67-57-187.mcn.bellsouth.net)
16:27:43 Quit Strife89 (Ping timeout: 260 seconds)
16:33:13 Quit TheLemonMan (Quit: "It's now safe to turn off your computer.")
16:38:18 Join jhMikeS [0] (~jethead71@d192-24-173-177.try.wideopenwest.com)
16:55:56 Join xorly [0] (~xorly@193.85.203.185)
17:00
17:16:43 Quit xorly (Quit: WeeChat 1.8-dev)
17:19:55 Quit ShameOnYou (Quit: Textual IRC Client: www.textualapp.com)
17:21:43***Saving seen data "./dancer.seen"
17:39:43 Quit petur (Quit: Connection reset by beer)
17:45:24 Quit alexweissman (Remote host closed the connection)
17:54:12user890104__builtin: can i allocate memory dynamically instead of converting the plugin to an overlay one?
17:57:49 Join PurlingNayuki [0] (~Thunderbi@2001:da8:215:4ff:8b0:c85f:69f6:90c9)
17:58:45 Part elensil
18:00
18:12:40 Join ShameOnYou [0] (~SOY@ANantes-257-1-183-126.w2-1.abo.wanadoo.fr)
18:12:40 Quit ShameOnYou (Changing host)
18:12:40 Join ShameOnYou [0] (~SOY@unaffiliated/shameonyou)
18:13:00 Part ShameOnYou
18:18:42 Join lorenzo92_1 [0] (~lorenzo@host120-107-dynamic.17-79-r.retail.telecomitalia.it)
18:24:36 Quit jhMikeS (Ping timeout: 246 seconds)
18:35:39 Quit marex-cloud (Remote host closed the connection)
18:41:23 Join alexweissman [0] (~alexweiss@149-160-162-52.dhcp-bl.indiana.edu)
18:44:13 Join marex-cloud [0] (sid137234@gateway/web/irccloud.com/x-xumkfelkiucjwuqa)
18:55:29 Quit alucryd (Remote host closed the connection)
18:57:23 Join alucryd [0] (~quassel@archlinux/developer/alucryd)
18:58:38 Quit PurlingNayuki (Ping timeout: 260 seconds)
19:00
19:00:40 Quit alexweissman (Read error: Connection reset by peer)
19:00:42 Join alexweis_ [0] (~alexweiss@2001:18e8:2:28b6:1d8c:64b1:2b71:61dd)
19:06:48 Quit lorenzo92_1 (Ping timeout: 260 seconds)
19:07:01 Quit krabador (Remote host closed the connection)
19:20:30 Join MrZeus2 [0] (~MrZeus@2a02:c7f:7066:fb00:cd9e:d86:599f:aab1)
19:21:46***Saving seen data "./dancer.seen"
19:54:16 Join lorenzo92_1 [0] (~lorenzo@host120-107-dynamic.17-79-r.retail.telecomitalia.it)
20:00
20:03:13 Quit alexweis_ (Remote host closed the connection)
20:04:52 Join krabador [0] (~krabador@unaffiliated/krabador)
20:18:23 Part robertd1
20:18:34bluebrotherlorenzo92_1: g#1601 contains some compiled binaries. I guess them being in is not intentional?
20:18:36fs-bluebotGerrit review #1601 at http://gerrit.rockbox.org/r/1601 : Samsung YP-R0/YP-R1: rbutil support by Lorenzo Miori
20:29:48 Join paulk-collins [0] (~paulk@gagarine.paulk.fr)
20:47:04lorenzo92_1bluebrother: definitely ! thanks for pointing them out
20:47:16 Join alexweissman [0] (~alexweiss@149-160-162-52.dhcp-bl.indiana.edu)
21:00
21:05:35 Join JanC_ [0] (~janc@lugwv/member/JanC)
21:06:52 Nick JanC is now known as Guest75375 (~janc@lugwv/member/JanC)
21:06:52 Quit Guest75375 (Killed (weber.freenode.net (Nickname regained by services)))
21:06:53 Nick JanC_ is now known as JanC (~janc@lugwv/member/JanC)
21:16:01 Join TheLemonMan [0] (~root@irssi/staff/TheLemonMan)
21:21:47***Saving seen data "./dancer.seen"
21:48:21 Quit lorenzo92_1 (Ping timeout: 264 seconds)
22:00
22:15:46 Join Strife89 [0] (~quassel@adsl-98-80-186-185.mcn.bellsouth.net)
22:17:02 Quit Strife1989 (Ping timeout: 260 seconds)
22:18:35 Join Barlow [0] (~barlow@unaffiliated/barlow)
22:24:41 Quit alexweissman (Remote host closed the connection)
22:33:25 Join alexweissman [0] (~alexweiss@2001:18e8:2:28b6:b4f8:eb57:deb2:b88f)
22:38:29 Quit krabador (Remote host closed the connection)
23:00
23:01:47 Quit alexweissman (Remote host closed the connection)
23:13:33 Quit TheLemonMan (Quit: "It's now safe to turn off your computer.")
23:17:39 Join alexweissman [0] (~alexweiss@2001:18e8:2:28b6:b1a1:a8eb:3341:b2d5)
23:21:52***Saving seen data "./dancer.seen"
23:39:17 Join Strife1989 [0] (~quassel@adsl-98-80-192-189.mcn.bellsouth.net)
23:42:07 Quit Strife89 (Ping timeout: 240 seconds)
23:49:50 Quit [Saint] (Remote host closed the connection)
23:51:16 Join [Saint] [0] (~sinner@rockbox/staff/saint)

Previous day | Next day