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 2012-10-17

00:00:59***Saving seen data "./dancer.seen"
00:07:03 Quit pamaury (Ping timeout: 246 seconds)
00:09:09 Quit [Saint] (Ping timeout: 246 seconds)
00:09:26 Quit KiwiCam (Ping timeout: 246 seconds)
00:14:34 Quit bertrik (Ping timeout: 252 seconds)
00:17:15 Join [Saint] [0] (~quassel@rockbox/user/saint)
00:22:00 Join Scr0mple [0] (~Simon@
00:25:48 Quit Scr0mple (Client Quit)
00:26:25 Quit sentriz (Quit: Leaving)
00:29:02 Join KiwiCam [0] (~quassel@
00:42:46 Quit ender` (Quit: Because nothing says "I have faith in God" like 3 inches of bulletproof glass on popemobile.)
00:45:17 Join amayer [0] (
00:49:18 Join Provel [0] (
00:55:28 Join factor [0] (~factor@
00:57:39 Join Gallomimia [0] (
01:42:06 Quit the-kyle (Quit: Leaving.)
01:44:29 Join the-kyle [0] (
02:01:02***Saving seen data "./dancer.seen"
02:23:19 Join saratoga [0] (123e0cfc@gateway/web/freenode/ip.
02:23:45saratogascorche`: spam filter ate a bunch of people's posts (including mine)
02:25:16JdGordonill fix it
02:26:31 Quit factor (Read error: Connection reset by peer)
02:38:43 Quit Provel (Read error: Connection reset by peer)
02:43:02 Join factor [0] (~factor@
02:45:21 Join Provel [0] (
02:46:29 Quit Gallomimia (Quit: I am likely going to change locations)
02:48:45 Part pacovila
03:14:32 Join MasT3R [0] (
03:20:22 Quit ParkerR_ (*.net *.split)
03:20:22 Quit Unhelpful (*.net *.split)
03:20:23 Quit zu (*.net *.split)
03:21:45 Join ParkerR_ [0] (ParkerR@unaffiliated/parkerr)
03:21:45 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
03:21:45 Join zu [0] (
04:01:04***Saving seen data "./dancer.seen"
04:12:21 Join Gallomimia [0] (
04:13:37 Quit XavierGr (Ping timeout: 246 seconds)
04:30:10 Quit MasT3R (Quit: CGI:IRC (EOF))
04:30:36 Join TheSphinX^ [0] (
04:34:00 Quit TheSphinX_ (Ping timeout: 272 seconds)
04:41:36 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:41:36 Quit amiconn (Disconnected by services)
04:41:39 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:42:37 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:42:37 Quit pixelma (Disconnected by services)
04:42:39 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:56:20CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
04:56:20*[Saint] prods JdGordon
04:56:35JdGordonforums again?
04:56:40[Saint]Akismet hates me again... ;)
04:56:48JdGordonakismet hates everyone again
04:57:20JdGordonscorche`: wtf is wrong with akismet? is there anything better or is our config broken?
04:57:22[Saint]Is there some rule along the lines of "if a post is edited within X of being posted == spam"?
04:57:31[Saint]That seems to be what is catching me out.
04:58:06[Saint]I always post, then go "Ah, dammit...I meant to say...", and then *poof*, it flags that post.
04:59:03[Saint]I didn't realize what all the times in the recent past have had in common until just now, but that seems to be it. Or, soe odd coincidence.
04:59:22JdGordonit doesnt have any ui in the forum i can see
05:02:11 Quit bootljhfdsds (Ping timeout: 260 seconds)
05:04:27*[Saint] giggles at the wiki thinking anything in CamelCaseis a topic
05:04:43[Saint]makes the credits look weird in the release notes.
05:06:25[Saint]"Johnny Johnson McSomenamehere" gets the "create this topic?" ? appended to it. And things like OpenSPC
05:11:34 Quit [7] (Disconnected by services)
05:11:42 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:28:54*[Saint] wonders how "%?ca<|>" is apparently valid.
05:29:25JdGordonwhy not?
05:29:31JdGordonits stupid, but valid
05:29:35[Saint]It's the abbreviated weekday tag.
05:30:03[Saint]%ca prints Sun, Mon, Tue depending on the day.
05:30:56JdGordonit will always do the empty line case
05:32:04[Saint]Ah, A guy in the forums tried: %?ca<%al%?cf<%cH:%cM|%cl:%cM>|00:00> which I interpreted as "check the abbreviated weekday, then check the 12/24hr time display preference and displays the correctly formatted time if it is a Sunday, and displays 00:00 if it is a Monday, and ignore all other days of the week"
05:34:14[Saint]When I read it, I believe he wanted %?cc (is there RTC?). It surprised me the above was valid.
06:01:05***Saving seen data "./dancer.seen"
06:46:45 Part amayer
07:02:11 Quit sciopa (Ping timeout: 248 seconds)
07:04:10 Quit Rower85 (Read error: Connection reset by peer)
07:24:01 Join bootljhfdsds [0] (~Prmhfhfx@
07:33:21 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
07:47:03 Quit kevku (Read error: Connection reset by peer)
07:47:39 Join mortalis [0] (
07:56:42 Quit Belzebub (Ping timeout: 260 seconds)
08:00:26 Quit Gallomimia (Quit: Gallomimia)
08:01:09***Saving seen data "./dancer.seen"
08:04:32 Join Belzebub [0] (
08:05:02 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:07:23 Join ender` [0] (
08:25:47 Join Jed`` [0] (~Jed@
08:25:55 Quit Jed`` (Client Quit)
08:28:17 Join LinusN [0] (
09:07:02 Join wodz [0] (
09:13:12mortaliswodz: turned out that bug with "no partition found" affects all hm601. It appeared after 71f7011 - 78ca74a
09:14:31mortalisi guess it another mysterious bug like with cache.
09:16:48wodzheh, this commits doesn't touch anything important so yeah it must be something fundamental
09:17:23wodzbtw. mini2440 doesn't build anymore
09:23:05wodzfirmware/SOURCES is borked
09:27:32Zagori whole bunch of targets fail building since the opus iram commit
09:27:43Zagor"region `PLUGIN_IRAM' overflowed by 640 bytes"
09:28:04ZagorI need to get build result announcements running again
09:29:42wodzI have a feeling that mini2440 was broken loooooooong ago
09:30:35Zagornot unlikely. targets that are not auto-built fall by the side pretty fast.
09:34:14wodzok, fixed
09:34:48 Join JdGord [0] (~AndChat80@
09:36:21[Saint]Is there an example anywhere of a svnversion->gerrit_magic adaption?
09:37:05[Saint]If I know what I'm supposed to do, fixing up ipodpatcher shouldn't be hard, but I'm not sure where to look for an example.
09:39:10[Saint]svnverion pops up in a few places still, it seems.
09:39:55wodz[Saint]: is your friend
09:41:51[Saint]wodz: aha, thanks.
09:43:04*[Saint] wonders how that (ipodpatcher) went unnoticed. Doesn't RbUtil require it?
09:43:24[Saint](perhaps it already existed in the tree the RbUtil releases are built from?)
09:43:31wodzit probably runs it as external tool or something
09:46:14mortalisno, rbutil use it as library
09:46:48 Join pamaury [0] (~quassel@
09:46:49 Quit pamaury (Changing host)
09:46:49 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:47:20mortalisand it builds for me, for linux at least
09:49:28 Quit JdGord (Read error: Connection reset by peer)
09:55:47[Saint]mortalis: it builds, yes. I should've been more explicit.
09:56:02[Saint]It compiles, but it throws a whole lot of errors.
10:01:10***Saving seen data "./dancer.seen"
10:09:49kugeldid anyone realize that there are many reds ?ß
10:10:18wodzkugel: Zagor pointed it out earlier today
10:14:45wodzingenic_jz47xx/crt0.S is copyrighted as GPL v2 only
10:29:53 Quit pamaury (Ping timeout: 240 seconds)
10:30:00 Quit TBCOOL (Ping timeout: 245 seconds)
10:36:14mortaliswodz: I'm starting to believe that cache issue wasn't cache issue or it wasn't fixed. I guess "no partition" issue & cache issue is the same problem. For now rb works with nop inserted in save_playlist_screen, but I haven't tried to find out boundary where exactly nop insertion stops to help.
10:37:11wodzthis pretty much looks like more general alignment problem
10:43:15 Join TBCOOL [0] (
10:45:09 Quit Provel (Read error: Connection reset by peer)
10:45:31 Join Provel [0] (
10:49:26 Quit TBCOOL (Ping timeout: 260 seconds)
10:55:18 Join TBCOOL [0] (
11:03:41 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:14:09 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
11:20:42wodzhow is called the file holding mp3 codec on the target?
11:32:00 Join XavierGr [0] (XavierGr@rockbox/staff/XavierGr)
11:38:52wodzI feel dumb but none of *.codec files reminds me of mp3 format
11:39:30gevaertsmpa.codec, I think
11:39:35gevaertsFor mpeg audio
11:39:46gevaerts(it also does mp2)
11:41:52wodzthanks, I'will never guess that :-)
11:55:01 Quit fs-bluebot (Ping timeout: 246 seconds)
11:56:20 Join fs-bluebot [0] (
11:56:40 Quit bluebrother (Ping timeout: 245 seconds)
11:58:36 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
12:01:11***Saving seen data "./dancer.seen"
12:03:37 Join swilde [0] (
12:29:46 Quit funman (Ping timeout: 240 seconds)
12:29:59 Join funman [0] (~fun@rockbox/developer/funman)
12:51:12 Quit mgottschlag (Disconnected by services)
12:54:47 Join petur [0] (~petur@rockbox/developer/petur)
12:56:13 Join kevku [0] (
13:00:03 Quit TBCOOL (Ping timeout: 248 seconds)
13:03:24wodzI'd like to ask for advice. elf_open() when used to load plugins needs to call audio_stop() before it starts to copy iram stuff. In case of codecs it can't do that for obvious reason.
13:03:58wodzAnyone have an idea how to resolve this other than passing additional arg to elf_open()?
13:09:10wodzhmm, I could check signature in header as it is always at the begining but thats rater hacky
13:09:51*[Saint] doesn't suggest that, then
13:09:55[Saint](was just about to)
13:12:39wodzmaybe it is not that bad if not only I thought about it.
13:13:21mortalisyou could check filename
13:13:44[Saint]It's definitely a little hacky, but, it's fairly trivial and would (read: _should_) "Just Work".
13:15:36wodzmortalis: Yes , that is another option but I much more believe magic byte sequence in mem then filename/extension
13:15:52wodz*believe in
13:19:14 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
13:30:52wodzworking :-) and it was just one line of code
13:39:19wodzhmm, plugins crashes if music is playing so I guess the check is not quite right
13:42:33 Quit mgottschlag (Ping timeout: 240 seconds)
13:53:31kugelwodz: how does it work currently?
13:55:04wodzkugel: what you mean?
13:57:51wodzif you are asking what I changed: on top of V12
14:01:15***Saving seen data "./dancer.seen"
14:12:38kugelwodz: I mean how it works in git master
14:13:49kugelah, i see. iram copy is handled in {codec,plugin}_crt0.c, so within the loaded binary. the elf loader needs to do it beforehand
14:15:27 Join amayer_ [0] (
14:15:38kugelwodz: why is commit_discard_idcache() needed in codec_load_ram()?
14:17:32wodzkugel: it is usually a good practice to commit caches before jumping into just loaded code. I was a bit surprised cache commit was handled in plugins/codecs crt0.c
14:18:01kugelwodz: wodz it's handled in lc_open()
14:18:17kugelit's handled in _crt0.c only because of the additional iram copy
14:18:24wodzI don't use lc_open()
14:18:47kugelright, i was talking about git master
14:19:12kugelanyway, it would be sufficient to call it in the loader after relocation
14:19:48wodzprobably, I'll change that
14:20:35kugelah, okay. i asked because i thought you'd already do it there and call it another time in codec_load_ram()
14:33:10 Join TBCOOL [0] (
15:01:18 Quit factor (Quit: Leaving)
15:12:55 Quit wodz (Quit: Leaving)
15:18:33 Join WalkGood [0] (~4@unaffiliated/walkgood)
15:18:48 Join einhirn [0] (
15:26:23 Part LinusN
15:40:57amayer_can someone unspam me?
15:42:55amayer_(my message was flagged as spam)
15:44:05gevaertsSomeone needs to tell akismet that we trust
15:44:05amayer_could we just use a captcha instead of a spam filter for the forums? have we talked about this before?
15:44:10gevaertsWe do
15:44:16amayer_gevaerts: agreed
15:44:23gevaertsIt doesn't stop many spammers
15:44:42amayer_where is it? i dont think i had to fill any out to post on the forums
15:45:03gevaertsThere should be one when registering
15:45:43the-kyleSpammers know how to get around captchas fairly easily. I was an admin on a forum and had to manually kill the spammers who got around the registration captcha.
15:46:03the-kyleSad but true.
15:46:40 Quit mortalis (Quit: Leaving)
15:47:08the-kyleThe forum I was on didn't have a post captcha, but if spammers weren't stopped at registration, they wouldn't have been stopped on an individual post either.
15:47:55*the-kyle was killing about 3 to 4 spam accounts per day at one point, and this was a fairly small community forum for a hosting site.
15:50:10amayer_wow... we have reCaptcha, a field to enter a value("no spam") and a filed to leave blank... its pretty crazy if a spammer gets through that
15:51:20amayer_unless they are manually doing it
15:51:50the-kyleWow you would think that would be enough, but apparently not. The hosting forum I was on didn't have that many roadblocks. But I guess spammers can get around just about anything these days.
15:52:53*nrg hangs head in shame
15:53:30Tornewe're pretty sure they are manually doing it
15:54:03Torneor at least, one guy did it manually one time and told everyone what to do :p
15:54:15the-kyleThis would explain how they pass the human tests. They are human, and put their bot on once the account is registered.
15:54:45Tornewell, for a given forum, as long as you have a reasonably successful way of breaking captchas, you only need to wokr out all the forum specific stuff once
15:54:57Torneafter that you can register more accouns with a bot exactly as successfully as you can solve captchas.
15:56:20the-kyleIt seems computers have gotten pretty good at solving captchas these days, which all but eliminates one obstacle for the spam bots.
15:56:39 Join lebellium_gs2 [0] (
15:56:40 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
15:56:46Tornewell, the problem is not really that simple :/
15:57:02Torneyou don't have to bevery good at solving captchas to make it worth it
15:57:16Torneespecially if you have a lot of IP addresses in different networks to use to do it from.
15:57:51the-kyleThis is true. Then you only need a fairly low success rate to setup spam accounts.
16:01:17***Saving seen data "./dancer.seen"
16:02:13 Quit Zagor (Quit: Clint excited)
16:02:18 Quit [Saint] (Read error: Connection reset by peer)
16:03:36the-kyleIt seems worth having a few false positives in exchange for keeping the amount of unblocked spam under control, as long as there are tweaks that can be done to keep the amount of false positives down as well.
16:04:23 Join [Saint] [0] (~quassel@rockbox/user/saint)
16:05:27Tornethe-kyle: i don't know that we *can* easily tweak akismet's behaviour here
16:05:33Torneit's a bit of a black box
16:07:27the-kylePerhaps there is a more tweakable spam filter available. Maybe something that works similar to e-mail filters, so the rate of false positives goes down over time.
16:12:26the-kyleI would do more to help, but unfortunately, I'm not very familiar with SMF. I was an admin on a MyBB forum, and it was quite a bit smaller than the Rockbox forum.
16:18:58 Quit lebellium_gs2 (Remote host closed the connection)
16:22:29 Join lebellium [0] (
16:25:15amayer_have we considered moving to another forum with built in features like we are talking about?
16:29:03 Join dfkt [0] (dfkt@unaffiliated/dfkt)
16:34:17 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
16:38:11 Join pretty_function [0] (~sigBART@
16:40:43 Quit Clear_runway (Ping timeout: 256 seconds)
16:52:22 Join Rower85 [0] (
17:11:49 Quit petur (Quit: *plop*)
17:13:40 Quit saratoga (Quit: Page closed)
17:58:20 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
18:01:21***Saving seen data "./dancer.seen"
18:07:49 Quit pretty_function (Ping timeout: 272 seconds)
18:15:25 Quit WalkGood (Quit: ♪ ♫ ♪ ♫ ♪ ♫ ♪)
18:59:47scorche|shwell, i have been meaning to upgrade to SMF 2.x at least
19:00:20 Join nateloaf [0] (
19:00:58scorche|shi should try adjusting the verification questions a bit and see if there is a brief respite from spam
19:01:16 Join WalkGood [0] (~4@unaffiliated/walkgood)
19:01:18scorche|shthe-kyle: yeah - there is absolutely no configuration we can do with akismet
19:02:12scorche|shi *think* akismet offers some method to tune its algorithms a bit, but that piece is not implemented in the SMF plugin used
19:03:50gevaertsI can't remember now, but I think we don't ever tell akismet about false positives
19:04:06 Join lebellium_gs2 [0] (
19:04:16 Quit lebellium_gs2 (Client Quit)
19:06:27 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
19:07:18gevaertsI wouldn't be surprised if that's why we're seeing more and more false positives lately
19:08:23 Join LinusN [0] (
19:27:06 Quit mgottschlag (Ping timeout: 246 seconds)
19:29:44[Saint]scorche|sh: my recipe to trigger the spam filter seems to be: post, then edit post within a few minutes of posting.
19:31:32[Saint]it'd be kinda nice if a post count higher than foo made one immune to such shenanigans.
19:33:51 Quit AlexP (Ping timeout: 245 seconds)
19:34:25 Join AlexP [0] (~alex@rockbox/staff/AlexP)
19:39:43amayer_[Saint]: that makes sense
19:40:05amayer_an accepted post count(not including mesages marked as spam)
19:40:59 Join Gallomimia [0] (
19:42:16 Part eckoit
19:42:52gevaertsSure, lots of things would be nice
19:43:19gevaertsImplementing them in the current forum software is a lot of work though
19:45:03gevaertsSMF1 was written in a time where "spam problem" meant a dozen or so spam posts per week
19:46:46the-kyleAn upgrade to SMF 2.x would probably be nice also, but there is the problem with database compatibility, meaning unless the devs made sure to make the SMF 1.x databases compatible with 2.x or unless they developed a conversion tool, it could be extremely difficult to upgrade.
19:48:25 Quit pamaury (Read error: Operation timed out)
19:51:32scorche|shthe-kyle: as far as i am aware, it shouldnt be too difficult to upgrade
19:52:12the-kyleIn that case, it could solve some problems, as long as the available spam filter plugins are configurable.
19:52:27scorche|shwhen i get a chance, i will see about spinning up another server, mirroring the current one, and testing the upgrade out there
19:53:44 Join webguest26 [0] (
20:01:25***Saving seen data "./dancer.seen"
20:04:08 Quit webguest26 (Quit: CGI:IRC (Ping timeout))
20:24:54 Join n1s [0] (
20:24:54 Quit n1s (Changing host)
20:24:54 Join n1s [0] (~n1s@rockbox/developer/n1s)
20:26:30 Join Horscht [0] (
20:26:30 Quit Horscht (Changing host)
20:26:30 Join Horscht [0] (~Horscht@xbmc/user/horscht)
20:28:29n1sah i forgot about the pp's with small iram, will fix shortly
21:08:19 Join sentriz [0] (~Senan@
21:12:50 Quit akaWolf (Ping timeout: 260 seconds)
21:20:23 Join eyfour [0] (
21:24:21 Join scorche [0] (~scorche@rockbox/administrator/scorche)
21:26:26 Quit scorche` (Ping timeout: 244 seconds)
21:27:09 Quit WalkGood (Quit: ♪ ♫ ♪ ♫ ♪ ♫ ♪)
21:32:47 Join pretty_function [0] (~sigBART@
21:32:49 Quit Rower85 (Read error: Connection reset by peer)
21:34:26 Part eyfour ("WeeChat 0.3.7")
21:39:51amayer_do we not have the commit bot anymore?
21:40:20 Join lovasoa [0] (~Thunderbi@2a01:e35:8a2e:8080:fde0:2f0c:a747:8e08)
21:41:14gevaertsThe CIA service shut down, and we don't have a replacement yet
21:41:56lovasoaI wrote a little patch to implement exponential and logarithm in the calculator plugin. Someone to review it ?
21:41:56lovasoa guessing that means we dont have source
21:42:34amayer_(for CIA) what does CIA stand for?
21:42:37amayer_was it just a perl script or something/
21:42:59gevaertslovasoa: any chance of putting it on gerrit?
21:43:10lovasoaWhat is gerrit?
21:43:15gevaertsamayer_: no, it was an external service used by lots of projects
21:43:29 Quit shamus (Read error: Connection reset by peer)
21:44:07gevaertslovasoa: see
21:44:22 Join shamus [0] (
21:44:25gevaertsWe're trying to get rid of patches on flyspray
21:45:20amayer_gevaerts: was it a gerrit plugin or something
21:46:06gevaertsamayer_: no, it was an external service that could be asked to monitor various VCS repositories
21:46:58amayer_oh ok.
21:47:00gevaertsWe didn't run it, we just used it
21:47:07amayer_i get it
21:47:25amayer_for an external service it was pretty quick
21:47:32amayer_...i kinda miss it
21:48:06gevaertsWell, we did have a gerrit hook to give it a nudge
21:48:20amayer_oh ok
21:55:18n1sit was very convenient and makes you forget to check the results manually
21:57:19amayer_i know right
21:57:31kugelcan't bluebrother's bot or logbot_ jump in?
21:57:54gevaertsLogbot would be easiest, since it runs on the same machine
21:59:05amayer_i think for the sake of keeping with the unix standard(keep tools simple and to 1 specific task) it should be a different bot
22:01:28***Saving seen data "./dancer.seen"
22:03:14amayer_im not sure how the bots work. but if its just a script or something if you keep them seperate you can remove or replace one very easy
22:03:27 Quit pretty_function (Ping timeout: 252 seconds)
22:03:48lovasoagevaerts: I get : To ssh://
22:03:48lovasoa ! [remote rejected] master -> master (prohibited by Gerrit)
22:03:48lovasoaerror: failed to push some refs to 'ssh://'
22:04:26 Quit y4n (Quit: We're fucking 3LN!)
22:04:27amayer_if you just installed ssh or git you might need to reboot
22:04:44AlexPamayer_: No you don't
22:04:45n1slovasoa: did you set up the ssh key on gerrit?
22:04:56amayer_i did when i had a fresh install
22:04:58AlexPamayer_: Please don't give random advice
22:04:58gevaertslovasoa: have a look at
22:05:19amayer_AlexP: im speaking from personal experiance
22:05:27AlexPYou're wrong :)
22:05:35amayer_i couldnt connect to gerrit at all. i had the ssh key and everything
22:05:43amayer_after logout and login it connected fine
22:05:49amayer_its in the log here somewhere
22:05:59AlexPWell that is very different to reboot
22:06:07amayer_i think it was a convo with [Saint]
22:06:14AlexPrestarting the session / new session is possible I guess
22:07:06amayer_AlexP: i wasnt assuming he was running Linux/Unix.
22:07:09amayer_if it was windows logout login doesnt do anything so it would need restart
22:07:29gevaertsFirst of all, that symptom didn't look like a failure to connect (more like "don't try to push to master if you want to upload a patch for review"), secondly, logout/login has nothing to do with rebooting, and thirdly, I'm still pretty sure there was a less invasive solution
22:08:39*AlexP cba
22:08:58amayer_oh gerrit is saying his push was rejected. i misread and thought it was saying his connection was rejected
22:09:22amayer_(which is the problem i was refering to)
22:09:24amayer_*bows* i appologize for the confusion
22:09:38n1scan we set custom errors for gerrit so it could say something like what gevaerts said?
22:10:17 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
22:10:39 Quit DarthDex ()
22:20:55lovasoaHere it is, I finally managed to set up gerrit:
22:21:26lovasoaBut I prefered the old svn and FS, it was way more simple!
22:22:26n1sgerrit is much nicer for reviewing
22:23:11lovasoaSo, will you review my patch?
22:23:22gevaertsYes. For instance it shows straight away that you have some tabs that should be spaces :)
22:24:00lovasoaSorry, my text editor is usuallu configured with
22:24:34lovasoatabs, and I forgot to change it before editing rockbox code...
22:24:45gevaertsAlso, gerrit only needs to be setup once. After that, it's a lot easier to push stuff to than FS
22:27:05gevaertsTo fix the tabs, you'd just edit the file, then git commit −−amend -a, and then git push origin HEAD:refs/for/master again
22:28:21 Quit stripwax (Read error: Connection reset by peer)
22:29:01lovasoagevaerts: Done !
22:42:25 Part LinusN
22:57:50 Join wodz [0] (
22:58:46 Quit zoktar (Ping timeout: 260 seconds)
23:00:47wodzelf codecs and plugins work perfectly ok on CF. On nano2g loading plugin when codec is loaded leads to undefined instruction :/
23:01:36 Quit tjb0607 (Excess Flood)
23:02:01 Join tjb0607 [0] (~tjb0607@
23:08:56 Join speckmade1 [0] (
23:12:42 Join factor [0] (~factor@
23:17:02speckmade1Recently I was kinda shocked about my Clip+ seeming to run his battery dry with less than 10 minutes of recording.
23:17:02speckmade1Assuming it may be the process of writing to the flash memory may be very consuming here, I radically turned down the sampling rate and got a whole lot of an improvement.
23:17:02speckmade1Can somebody confirm that writing to flash is the most consuming thing here? Can maybe somebody point me to some numbers?
23:17:02DBUGEnqueued KICK speckmade1
23:17:02speckmade1How much heavier is it on the battery than running the MP3 encoder? I.e. can I save some battery life and yet retain more quality by saving to MP3?
23:17:43Torneless than ten minues seems pretty terrible, but i would guess that wrting to the flash is the expensie part, yes
23:17:53Tornei doubt anyone has runt hat experiment on that device, though, you'll have to time it for yourself
23:19:18speckmade1so it could be useful if I did some benchmarking, eh?.. :-)
23:21:07wodzthere was article on /. recently about recording on AMSv2 with rb (it was clipv2 or clip+ can't remember) and they recorded for a few hours on single charge. From my calc the limiting factor was player's flash capacity not the battery I recall.
23:21:57speckmade1wooops - so it may be my experimental firmware?..
23:22:03 Join pamaury [0] (
23:22:03 Quit pamaury (Changing host)
23:22:03 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:22:32wodzthey used wav recording - maybe this matters
23:22:33speckmade1I have a nightly build running here...
23:23:31speckmade1First I used WavPack and it seemed to draw a lot of power. Then I switched to RIFF-WAVE - seemed to be the same thing.
23:24:39speckmade1Now when I reduced the sampling rate from 48 k down to 8 and removed the useless second channel I don't see remarkable battery drain anymore.
23:25:41wodzdunno, it just ringed the bell
23:27:00speckmade1With WAV, 48 k, stereo I remember something like 6 minutes of recording with a battery that was well above 70 % charged, I think, and it was down below 10 % or so.
23:29:07 Nick ParkerR_ is now known as PorkerR (ParkerR@unaffiliated/parkerr)
23:31:11 Quit lovasoa (Quit: lovasoa)
23:31:25 Quit amayer_ (Quit: going ~/)
23:32:25 Quit wodz (Quit: Leaving)
23:34:03 Quit Horscht (Quit: Verlassend)
23:36:38 Quit mystica555 (Ping timeout: 244 seconds)
23:43:22 Quit AlexP (Ping timeout: 260 seconds)
23:44:02 Join AlexP [0] (~alex@rockbox/staff/AlexP)
23:45:07 Join amayer_ [0] (
23:46:51speckmade1They mentioned 48 k sampling rate being buggy and recommend 44.1 k. That sounds like a useful hint. Thanks, wodz!
23:49:27 Quit amayer_ (Client Quit)
23:51:33 Quit n1s (Quit: Ex-Chat)
23:53:58 Join Clear_runway [0] (~Clear_run@

Previous day | Next day