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 2020-08-12

00:00:16 Join TheEaterOfSouls [0] (~souls@unaffiliated/theeaterofsouls)
00:29:42***Saving seen data "./dancer.seen"
00:41:11__builtinit's nice to see an uptick in activity recently :)
00:41:26__builtinperhaps we should start thinking about our release strategy going forwards
00:42:17__builtinpersonally I'm in favor of moving to a rolling model, but there is still something to be said about having a "known good" release available
00:59:21 Join Rower [0] (
01:55:08 Quit ac_laptop (Ping timeout: 256 seconds)
02:29:43***Saving seen data "./dancer.seen"
03:16:55 Join petur [0] (~petur@
03:16:55 Quit petur (Changing host)
03:16:55 Join petur [0] (~petur@rockbox/developer/petur)
03:58:10 Quit E14n (Ping timeout: 256 seconds)
04:11:49 Join E14n [0] (
04:29:44***Saving seen data "./dancer.seen"
04:59:25 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
05:05:39 Quit danielp3344 (Quit: killed)
05:05:40 Quit kadoban (Quit: killed)
05:05:46 Quit blbro[m] (Quit: killed)
05:09:13 Quit pamaury (Ping timeout: 246 seconds)
05:09:26 Quit efqw (Quit: Connection closed for inactivity)
05:13:35 Join kadoban [0] (kadobanmat@gateway/shell/
05:40:14 Join danielp3344 [0] (danielp334@gateway/shell/
05:40:20 Join blbro[m] [0] (blbrostrat@gateway/shell/
05:46:45 Quit paulk-leonov (Ping timeout: 240 seconds)
05:48:10 Join paulk-leonov [0] (
05:54:54 Quit E14n (Ping timeout: 256 seconds)
06:29:48***Saving seen data "./dancer.seen"
06:32:31 Quit St3ak (Read error: Connection reset by peer)
06:32:46 Join St3ak [0] (
06:43:05 Join Soap_ [0] (~Soap@rockbox/staff/soap)
06:46:21 Quit Soap (Ping timeout: 265 seconds)
06:50:00 Join E14n [0] (
06:58:11 Join Soap [0] (~Soap@rockbox/staff/soap)
06:59:45 Quit Soap_ (Ping timeout: 240 seconds)
07:13:25 Quit E14n (Ping timeout: 264 seconds)
07:27:16 Join E14n [0] (
08:29:52***Saving seen data "./dancer.seen"
08:42:00 Join massiveH [0] (
09:10:32speachy__builtin: what, every 6 or 12 months?
09:11:29speachytbh without a more concerted testing effort for the back catalog I don't know that "releases" will be any more stable than the rolling stuff we're doing now..
09:12:47speachythat said a release requires quite a few manual steps right now. If we get the "release cost" down through better tooling doing it more often wouldn't be a bad thing..
09:14:23speachyhmm. I should probably update the MajorChanges page to cover the past month. Including such minor details like the sh purge.
09:39:57speachyhmm, the jz7460 reclocking changes made shutdowns a little unstable.
09:40:58speachyOOTH, until about three weeks ago ROLO didn't work at all, so it's still a net gain
09:56:47 Join __Bilgus_ [0] (41ba23be@
09:57:10__Bilgus_speachy what doyou mean by unstable?
09:57:36speachyhanging on the "shutting down" splash
09:57:55speachy(or ROLO... screen)
09:58:18__Bilgus_have a look at the cfg file after these events and see if its very large
09:59:05__Bilgus_there was this weird bug that caused 60mb cfg files due to a little less than strict file write function
10:02:26 Join ac_laptop [0] (~ac_laptop@
10:02:56__Bilgus_oh nm I guess I already fixed that in g#1975
10:02:58fs-bluebotGerrit review #1975 at : Fix vuprintf fix possible %s buffer over-read by William Wilgus
10:03:26speachythe most recent time was this morning, and it also trashed the recent bookmarks list
10:04:03speachyI'd guessimate about 1 in 5 shutdowns/ROLOs is screwing up
10:04:18speachymost (but not all) were playing at the time.
10:07:06__Bilgus_rolo takes over the core buffer, I could see there being something taking issue with moving ram around
10:13:01speachyI'm not sure if ROLO stability has anything to do with the shutdown issues; I didn't even realize it worked until last week. :)
10:16:07 Join johnb2 [0] (
10:17:23__Bilgus_you might try disabling the config save code and see if it helps
10:17:44 Quit massiveH (Quit: Leaving)
10:19:23johnb2__Bilgus_: I got me another Sansa e200v1 and wanted to do a battery benchmark. With the current dev build, it just hangs if nothing is playing, but a "Prefetch abort ..." if there was a paused playback.
10:20:10johnb2It works fine with an older dev build, so I am suspecting this might be due to your recent battery_bench changes.
10:20:13speachyjohnb2, I wasn't able to figure out your gerrit IDs based on what you sent me. Can you send me the userid for the new one, at least?
10:22:17__Bilgus_johnb2 can you try the announce plugin it uses the same TSR/thread buffer code
10:22:30__Bilgus_its under demos
10:29:50__Bilgus_maybe we should size the sample buffer smaller depending on the size of the plugin buffer it reeks of stack overflow to me
10:29:53***Saving seen data "./dancer.seen"
10:42:43 Join johnb3 [0] (
10:44:31 Quit johnb2 (Ping timeout: 246 seconds)
10:45:55johnb3__Bilgus_: what is the announce plugin supposed to do? Playback continues and the player is responsive.
10:51:36 Join johnb4 [0] (
10:54:16 Quit johnb3 (Ping timeout: 256 seconds)
11:01:22__Bilgus_after you start it it runs in the background if you open it again it'll announce the items you set-up
11:02:02__Bilgus_it'll eventually be tied to the wps hotkey when I get this other code finished
11:03:18__Bilgus_So each time you re-run the announce plugin it does the same task switch Battery Bench does every 60 seconds
11:04:32__Bilgus_if that crashes then I need to closely look at the thread stack maybe something fundamentally wrong with my implementation if it doesn't then need to start looking very closely at BatteryBench
11:06:16johnb4assuming I have the .voice file installed?
11:06:46__Bilgus_i'm thinking it'd not announce without the voice file but it'd still run
11:06:56__Bilgus_let me double check
11:07:37johnb4yes, on first run I see a message 'hold for settings'. Let me install a voice file.
11:08:28__Bilgus_it'll still work without it
11:08:50__Bilgus_and it defaults to the time and date if you don't configure anything
11:09:17 Quit inkblot_sandwich (Ping timeout: 244 seconds)
11:09:52__Bilgus_and I should probably put in a message about no voice and quit the plugin rather than having it consuming resources for no good reason
11:10:10__Bilgus_but in this case it'll do :p
11:15:10johnb4yes, it is announcing time now.
11:15:17johnb4so this plugin is working.
11:15:50__Bilgus_hmm as much as I hate to say it I hoped it'd crash
11:16:53__Bilgus_there is something going on with battery bench its not doing any disk writes so maybe one of the pieces of information it is gathering
11:17:20__Bilgus_it does do that first one to write out the header
11:17:41__Bilgus_I imagine thats done before it goes TSR
11:18:58__Bilgus_johnb4 do you get the empty battery bench file or your disk?
11:19:12__Bilgus_well the empty header stuff..
11:19:23johnb4let me check
11:21:04johnb4# −−File already present. Resuming Benchmark−−
11:21:04johnb4# Battery bench run for Sandisk Sansa e200 series version b4e70422a3-200811
11:21:04johnb4# Rockbox has been running for 00:02:59
11:21:15johnb4so, yes.
11:21:43__Bilgus_ok so that part is good its written while the plugin is still in the foreground
11:22:30johnb4I will be back later
11:22:58__Bilgus_ok ill give you a few things to try then..
11:29:19__Bilgus_Try #1 don't record the data just move through the data buffer
11:29:43 Quit __Bilgus_ (Remote host closed the connection)
11:30:16 Join __Bilgus_ [0] (41ba23be@
11:31:22__Bilgus_and the other try might be try removing SHAREDBSS_ATTR from the even queue @328
11:31:34__Bilgus_event queue*
11:34:50 Quit petur (Read error: Connection reset by peer)
11:47:03 Quit Rower ()
11:51:33 Join Rower [0] (
11:52:45 Join advcomp2019 [0] (
11:52:45 Quit advcomp2019 (Changing host)
11:52:46 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
11:54:40 Quit advcomp2019_ (Ping timeout: 256 seconds)
12:04:21 Part flab ("bye")
12:19:22 Quit johnb4 (Ping timeout: 246 seconds)
12:23:30 Join advcomp2019_ [0] (
12:23:30 Quit advcomp2019_ (Changing host)
12:23:30 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
12:27:13 Quit advcomp2019 (Ping timeout: 264 seconds)
12:29:56***Saving seen data "./dancer.seen"
12:31:25 Quit Rower (Ping timeout: 264 seconds)
12:33:26 Quit __Bilgus_ (Remote host closed the connection)
12:57:46 Join __Bilgus_ [0] (41ba23be@
12:59:00__Bilgus_Open_plugins is just about done g#2623
12:59:03fs-bluebotGerrit review #2623 at : Add open_plugin to core by William Wilgus
12:59:47__Bilgus_still need to add some lang entries
13:02:32 Join johnb4 [0] (
13:03:54 Quit johnb4 (Client Quit)
13:08:00 Join johnb4 [0] (
13:08:59 Join lebellium [0] (
13:42:54 Join MrZeus_ [0] (
13:44:07 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
13:44:59 Join Rower [0] (
13:46:18 Quit johnb4 (Ping timeout: 256 seconds)
13:50:20 Join johnb4 [0] (
14:12:57 Join amiconn_ [0] (jens@rockbox/developer/amiconn)
14:12:58 Quit amiconn (Killed ( (Nickname regained by services)))
14:12:58 Nick amiconn_ is now known as amiconn (jens@rockbox/developer/amiconn)
14:20:45speachywww stuff is down
14:21:36speachylooks like a massive bot storm hitting the wiki
14:21:36 Quit kadoban (Write error: Connection reset by peer)
14:21:40 Quit danielp3344 (Remote host closed the connection)
14:21:40 Quit blbro[m] (Remote host closed the connection)
14:28:12 Join blbro[m] [0] (blbrostrat@gateway/shell/
14:29:58***Saving seen data "./dancer.seen"
14:32:59 Quit pamaury (Ping timeout: 240 seconds)
14:34:52speachysomething has seriously lost its mind
14:39:43 Quit johnb4 (Ping timeout: 246 seconds)
14:44:05 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
14:55:26 Join danielp3344 [0] (danielp334@gateway/shell/
14:55:26 Join kadoban [0] (kadobanmat@gateway/shell/
15:01:02Strife89Is there a tool that can scan for corrupted audio files? I've run fsck on my iPod's old hard drive (now in the iRiver), but it still hangs on some files and skips large chunks of others
15:01:54Strife89Which, incidentally, makes me more suspect of my iPod Color's logic board and the ribbon cable connecting a disk to it.
15:04:45 Quit tchan (Ping timeout: 240 seconds)
15:05:12speachywell, it's also possible that rockbox's decoders aren't as robust as they should be
15:06:47Strife89Perhaps, but several of the files I've tried played perfectly fine on other players (most notably, my Clip+ and Zip, which I can't remember ever experiencing such issues with).
15:11:56speachyserver's back up; turns out it was an internal user DoS rather than external.
15:17:46 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
15:25:23 Join johnb4 [0] (
15:29:20johnb4__Bilgus_: were you elaborating about your ideas @11:29 or was this meant as instructions for me (then I didn't get it :-)
15:35:05 Quit jdarnley (Ping timeout: 240 seconds)
15:35:35 Join J_Darnley [0] (
16:09:06 Quit johnb4 (Ping timeout: 256 seconds)
16:09:58 Join reductum [0] (
16:11:12 Join johnb4 [0] (
16:15:25 Quit johnb4 (Ping timeout: 240 seconds)
16:30:02***Saving seen data "./dancer.seen"
18:07:03 Quit lebellium (Quit: Leaving)
18:07:16 Quit kugel_ (Ping timeout: 246 seconds)
18:14:48 Join cockroach [0] (~blattodea@pdpc/supporter/active/cockroach)
18:18:32 Join advcomp2019__ [0] (
18:18:32 Quit advcomp2019__ (Changing host)
18:18:32 Join advcomp2019__ [0] (~advcomp20@unaffiliated/advcomp2019)
18:21:12 Quit pamaury (Ping timeout: 260 seconds)
18:21:23 Quit advcomp2019_ (Ping timeout: 240 seconds)
18:30:04***Saving seen data "./dancer.seen"
18:31:19 Join Oksana [0] (~Wikiwide@Maemo/community/ex-council/Wikiwide)
18:42:25 Quit reductum (Quit: WeeChat 2.9)
18:54:26 Quit ac_laptop (Ping timeout: 260 seconds)
18:59:51 Join efqw [0] (uid412670@gateway/web/
19:06:22 Quit amiconn (Ping timeout: 260 seconds)
19:06:57 Quit pixelma (Ping timeout: 260 seconds)
19:11:48 Join pixelma [0] (marianne@rockbox/staff/pixelma)
19:11:49 Join amiconn [0] (~jens@rockbox/developer/amiconn)
19:13:09__Bilgus_b4 instructions were prior
19:14:18__Bilgus_oh I guess in the logs yes 11:30 sorry
19:14:21__Bilgus___Bilgus_Try #1 don't record the data just move through the data buffer11:29__Bilgus_ the other try might be try removing SHAREDBSS_ATTR from the even queue @32811:31__Bilgus_event queue*
19:16:52 Quit amiconn (Ping timeout: 260 seconds)
19:16:52 Quit pixelma (Ping timeout: 260 seconds)
19:18:52 Join pixelma [0] (marianne@rockbox/staff/pixelma)
19:19:07 Join amiconn [0] (jens@rockbox/developer/amiconn)
20:22:13 Quit MrZeus_ (Ping timeout: 265 seconds)
20:22:13 Quit __Bilgus_ (Remote host closed the connection)
20:29:36 Join ac_laptop [0] (~ac_laptop@
20:30:09***Saving seen data "./dancer.seen"
20:38:35 Quit koniu (Remote host closed the connection)
20:39:25 Join koniu [0] (~koniu@gateway/tor-sasl/koniu)
20:47:55fs-bluebotBuild Server message: New build round started. Revision 4d9f190, 280 builds, 9 clients.
20:55:18 Join kugel [0] (
20:55:18 Quit kugel (Changing host)
20:55:18 Join kugel [0] (~kugel@rockbox/developer/kugel)
21:00:27 Quit emacsomancer (Ping timeout: 260 seconds)
21:05:42 Quit cockroach (Quit: leaving)
21:11:34 Join emacsomancer [0] (
21:11:57fs-bluebotBuild Server message: Build round completed after 1443 seconds.
21:12:01fs-bluebotBuild Server message: Revision 4d9f190 result: All green
21:12:02fs-bluebotBuild Server message: New build round started. Revision 7ed1263, 280 builds, 8 clients.
21:26:46 Join inkblot_sandwich [0] (
21:29:58fs-bluebotBuild Server message: Build round completed after 1076 seconds.
21:29:59fs-bluebotBuild Server message: Revision 7ed1263 result: All green
21:47:33 Join __BILGUS_ [0] (41ba23be@
21:56:53 Quit koniu (Remote host closed the connection)
21:57:18 Join koniu [0] (~koniu@gateway/tor-sasl/koniu)
22:18:16 Join flab [0] (
22:26:02 Quit efqw (Quit: Connection closed for inactivity)
22:30:11***Saving seen data "./dancer.seen"
22:36:13 Quit Oksana (Ping timeout: 264 seconds)
22:45:40 Quit ps-auxw (Disconnected by services)
22:45:49 Join ps-auxw [0] (
22:48:41 Part flab
22:54:05 Join Oksana [0] (~Wikiwide@Maemo/community/ex-council/Wikiwide)
23:07:44fs-bluebotBuild Server message: New build round started. Revision a9ac2d0, 280 builds, 9 clients.
23:18:22 Quit [7] (Ping timeout: 260 seconds)
23:18:40 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
23:19:50__builtinspeachy: you're right - the value of releases is diminished if we don't put extra testing effort into a release vs. the dev builds
23:21:04 Quit __BILGUS_ (Remote host closed the connection)
23:21:16fs-bluebotBuild Server message: Build round completed after 812 seconds.
23:21:21fs-bluebotBuild Server message: Revision a9ac2d0 result: All green
23:26:09speachyand it's not like any of this testing can be automated either.
23:27:10__builtinshort of something like the nanotron :)
23:30:54__builtinthough come to think of it, if we had a _lot__ of developer time we could actually pull that off
23:31:10__builtinwe could leverage the build farm infrastructure with hardware-attached builders
23:33:21__builtinsee if that sounds like gibberish
23:34:54speachyif we had a lot of developer time this conversation would be moot. :D
23:36:03__builtinthe alternative is a full-fledged emulator for some representative targets
23:36:09__builtinas opposed to a simulator as we currently have
23:37:36__builtinthere was some preliminary work on that back in the day:
23:38:51speachyemulation is pointless IMO.
23:39:26speachybecause the emulator has to be quirk-for-quirk compatible with the real hardware.
23:39:32__builtinit covers a large percentage of accidental, "woops, I broke the build" logic errors though
23:39:54__builtinI mean, if someone is pushing driver changes presumably they already have the device
23:40:27speachytrue enough.
23:40:52speachythe UIsim builds do a pretty good job at covering code above the driver layer
23:41:20__builtinperhaps we could implement some sort of unit testing framework on top of that
23:41:42__builtinhaving the ui sim does make full-on emulator moot...
23:42:13speachyperhaps... but in the end it's still divorced from the realities of the underlying hardware [&drivers]
23:42:35speachywhich is where most of our hair-pulling comes from
23:43:03__builtinyeah, I guess if a is visible in the UI sim then enough people will complain
23:43:07__builtin*a bug
23:43:47speachyheh heh
23:44:38__builtinmaybe we should just call that "post-release quality assurance"
23:44:41speachyjust occurred to me that writing the UI for the bluetooth features of the Rocker/etc means we could also do that on the UIsim
23:45:16speachyand shunt the audio through the host system's own bluetooth layer.
23:45:42speachy... that actually seems like a useful feature.

Previous day | Next day