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 2014-07-08

00:02:01 Quit Geoff_ (Ping timeout: 252 seconds)
00:04:47 Quit ender` (Quit: Debian comes in three flavours: stale, rusting and broken.)
00:10:20 Quit ZincAlloy (Quit: Leaving.)
00:27:55 Quit Misanthropos (Ping timeout: 264 seconds)
00:32:24 Quit bertrik (Remote host closed the connection)
00:45:10***Saving seen data "./dancer.seen"
01:04:19 Quit bluebrother (Ping timeout: 248 seconds)
01:06:28 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
01:19:55 Quit Topy44 (Quit: ZNC -
01:21:54 Join Topy44 [0] (
01:41:20[Saint]Can someone please refresh my memory as to how to get the SDL application to install in a chosen directory?
01:41:52[Saint]I can't remember the correct syntax to pass to 'make install'.
01:48:07 Quit bcobco ()
02:05:46 Quit bzed (Ping timeout: 252 seconds)
02:07:13 Join bzed [0] (
02:16:12 Join Strife89 [0] (
02:17:51[Saint]For some fucked up reason, "make install −−install=/home/saint/Development/Rockbox/SDL/480x800-20140708" is installing to /dev/null
02:26:25 Quit bzed (Ping timeout: 245 seconds)
02:28:17 Join bzed [0] (
02:28:19 Quit evilnick (Ping timeout: 252 seconds)
02:28:52 Quit gevaerts (Ping timeout: 252 seconds)
02:45:14***Saving seen data "./dancer.seen"
02:50:18 Join ygrek [0] (~user@
02:57:43 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
03:00:00 Quit AlexP (Remote host closed the connection)
03:01:43 Join AlexP [0] (~alex@rockbox/staff/AlexP)
03:01:56 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
03:51:25 Quit ygrek (Ping timeout: 245 seconds)
04:10:48 Join ygrek [0] (~user@
04:41:56 Quit amiconn (Disconnected by services)
04:41:57 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:41:58 Quit pixelma (Disconnected by services)
04:41:58 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:42:01 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:42:06 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:44:25 Part xiaopi[z]
04:44:51 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
04:45:18***Saving seen data "./dancer.seen"
04:59:48 Join xiaopi[z] [0] (~xiaopiz]
05:09:09 Quit steffengy (Disconnected by services)
05:09:10 Join steffengy1 [0] (
05:27:16 Nick DormantBrain is now known as SuperBrainAK (~andy@
05:33:31 Quit [7] (Ping timeout: 240 seconds)
05:34:53 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:47:40 Quit AKBlindy (Ping timeout: 240 seconds)
05:50:13 Join AKBlindy [0] (
05:50:33 Quit the-kyle (Ping timeout: 240 seconds)
05:51:33 Quit kuldeepdhaka (Ping timeout: 240 seconds)
05:52:29 Join the-kyle [0] (
06:01:51JdGordonjhMikeS: ping?
06:03:39JdGordonwanted to reply to your comments on gerrit but here is better
06:03:49JdGordonwhat are you storing in the dircache buffer as offsets?
06:04:36jhMikeSjust offsets into the buffer, either entry array indexes or string byte offsets
06:05:28 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
06:05:51JdGordonthe skin engine has somemhing like a dozen differnt struct types getting dumped in the buffer, i was hoping to try get back some compiler type checking going back to pointers
06:05:51jhMikeSbut the base buffer pointer is moved one entry behind where the buffer really starts
06:06:31JdGordonalso, did you do a binary size comparisson? I found a pretty big difference removing the offsets
06:06:50jhMikeSNot on your code, no
06:07:05JdGordonno, on yours
06:07:14JdGordonim seeing -5K or more
06:07:38JdGordonalso, considering the buffer doesnt actually move in practice the potentially slow management isnt an issue
06:07:48jhMikeSIt's not just the dircache. The code is much more functional. Dircache is twice as long since it's does so much more than it did.
06:08:29jhMikeSSo, it's not easy to compare binsize. What it does do for most targets is save actual memory
06:08:37JdGordonmoving the offsets so 0 is invalid would have been a good idea though
06:10:28JdGordonas for the crashing thing, I was hoping that any not updated pointers would cause a data abort instead of just pointing to rubbish which can afely be dereferenced - which i've found to be mostly true so far
06:10:58jhMikeSIt might or just scribble on something instead :D
06:11:39jhMikeSThe movement might not be enough to throw them out of valid memory locations
06:12:09JdGordonno, but zeroing out the temp buffr before it is copied is a pretty good start
06:12:14JdGordonyes, it isnt perfect :)
06:14:34jhMikeSbtw, my e200 binsize goes up 12K but memory usage is 3K less
06:15:46JdGordondoes dircache use 1 single buflib handle or multiple? and does it/they resize?
06:16:04jhMikeSjust one. it compacts after building
06:16:38jhMikeSIt always did that but takes liberties it couldn't before buflib
06:16:50JdGordonok, so it is a little bit similar there
06:17:07jhMikeSit just has to adjust offsets to the names since they're allocated from the end
06:17:28jhMikeSbut it also recycles freed stuff
06:18:08 Quit kuldeepdhaka (Read error: Connection reset by peer)
06:23:04jhMikeSit only adjust during a shrink, under its own control, not when moving. moving the buffer just changes one pointer.
06:23:52jhMikeSthough if you have offsets into one buffer from another buffer or something crazy... :)
06:24:14JdGordonit isnt that bad :)
06:24:55JdGordonhow hard would it be for you to try going back to pointers?
06:25:11JdGordondoes arm eembly suck at doing &buffer[offset]?
06:25:31jhMikeSnot at all
06:25:50jhMikeSI could try it, just for fun or something?
06:26:20JdGordonyeah, assuming we care about thing like code size still
06:27:11jhMikeSthere's one issue though, I'd bloat the entries. "up" indexes at the root level identify the volume number and they're negative
06:28:25jhMikeSand it would help much with names since they might be stored directly or indirectly so that needs checking when looking that up
06:30:40JdGordonbut on the topic, my intention is to make the skin engine more objecty with the callbacks struct each token thing has now to make things cleaner
06:30:55jhMikeSI have all of 26 get_entry calls to convert indexes to pointers in 3011 lines :\
06:30:57JdGordonif the pointer thing is really objected to i can think about not including it
06:31:13JdGordondont bother
06:37:16jhMikeSI don't know. I'm thinking about fancier threading so this thing can run directly on host OS threads and that stuff makes it much more difficult
06:37:39*JdGordon doesnt really need to worry about that
06:38:41jhMikeSbut buflib just won't work with that on target, not using like it is right now
06:40:12jhMikeSalot of stuff is easy to change and alot is already suitable, except huge modules like the database and buflib
06:43:49jhMikeSbuflib ought to just hand out malloc pointers on a host, make the handle type an intptr_t and drop the table altogether
06:45:22***Saving seen data "./dancer.seen"
06:45:58JdGordonas long as host != sim then yeah
06:47:19jhMikeSreally it should always be an app that just happens to come with it's own mini kernel for a native target to minimize the differences
06:47:53JdGordonis that not the firmware/ apps/ split?
06:48:18jhMikeSit doesn't feel very split though
06:48:29JdGordonbecause firmware/ is all over the place
06:48:37JdGordonthat needs to be split up again
06:48:50jhMikeSalot of stuff should be /lib
06:49:36jhMikeSmaybe a move to event-driven UI?
06:50:14jhMikeSgtk super-light
06:51:16JdGordongo for it!
06:51:40 Join goom [0] (
06:52:29jhMikeSI seriously probably will.
06:53:11jhMikeSI'd really like un-stovepipe this shit. Just have a message loop that exits on power off and does the shutdown in main
06:55:01jhMikeSthat would make it all like the app/sim code
06:57:30jhMikeSanyway, these big changes are for my own experiment. If noone wants them in the trunk that fine. I'll just work it out in my own branch for my amusment since that's why I bother anyway.
07:00:00jhMikeSAt this rate I'll need life-extension therapy
07:01:34jhMikeSbtw, were we taking HWCODEC out or not?
07:02:08JdGordoni believe the general concensus is it can go when someone does something to remove it
07:02:50jhMikeSI did in one of my branches, not absolutely everything, but the code is gone. it actually didn't take alot of time.
07:02:56 Join saratoga [0] (123e1e00@gateway/web/freenode/ip.
07:03:13jhMikeSthat an charcell
07:03:16saratogai believe the decision was that we would stop maintaining it after 3.14
07:04:44jhMikeSwas that said on IRC? I don't remember seeing more ML discussion on it (just a bunch of "me too")
07:13:44 Join kugel [0] (~kugel@rockbox/developer/kugel)
07:14:13*jhMikeS feels itchy to start deleting stuff
07:15:38saratogathen you have to do a release!
07:16:00*jhMikeS can realease all on his own :P
07:17:06saratogawe should probably see what happens with the recent USB work
07:17:14saratogaand if that goes in, branch soon thereafter
07:18:41kugeljhMikeS: i have a patch on gerrit which implements malloc-based buflib
07:19:14kugelthis is my solution to buflib being unworkable currently for preemtible
07:19:48kugel(the other one would be moving to a lock-unlock scheme i guess)
07:20:15kugelJdGordon: you still haven't answered my question
07:20:24jhMikeSI did have a couple works for that but on a host I see no reason to have all the callback stuff
07:21:13kugelthe callbacks can be mostly ignored, except in buflib_shrink
07:22:14jhMikeSwhy would that be needed?
07:24:17jhMikeSyou mean if you use realloc(), then the buffer needs adjusting
07:25:34 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
07:27:08kugelthe pointers can change with realloc
07:28:11jhMikeSexactly, I was thinking about why then realized
07:28:35*jhMikeS hasn't realloc'ed anything in years
07:28:42kugelbut _shrink is not problematic because it's explicitly called, not behind the back via some other allocation
07:29:19jhMikeSI did have an idea about threads calling their own callbacks
07:29:44jhMikeSthey wouldn't know it but it would be part of a lock/unlock scheme
07:30:17jhMikeSfirst one to lock it after a move
07:33:12jhMikeSyou can also guard the callback with a mutex test, if it can grab the lock, it can move things, if it fails to grab it, it can't move
07:41:35 Nick pookie_ is now known as olspookishmagus (
07:44:30 Join mortalis [0] (~kvirc@
07:47:53 Quit amiconn (Remote host closed the connection)
07:47:53 Quit pixelma (Remote host closed the connection)
07:49:15 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
07:49:17 Join amiconn [0] (amiconn@rockbox/developer/amiconn)
07:56:39 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
08:10:43 Quit pamaury (Ping timeout: 240 seconds)
08:17:31 Quit ygrek (Ping timeout: 240 seconds)
08:27:19 Join ender` [0] (
08:43:44 Join einhirn [0] (
08:45:26***Saving seen data "./dancer.seen"
09:08:14kugel_[Saint]: −−prefix=<dir> at configure time, as usual
09:09:23 Nick SuperBrainAK is now known as DormantBrain (~andy@
09:15:31 Join petur [0] (~petur@rockbox/developer/petur)
09:18:28 Join ygrek [0] (~user@
09:19:55 Quit l2k-Shadow (Ping timeout: 260 seconds)
09:23:31 Join l2k-Shadow [0] (
09:23:31 Quit l2k-Shadow (Client Quit)
09:45:03 Quit AlexP (Ping timeout: 240 seconds)
09:45:17 Join AlexP [0] (~alex@rockbox/staff/AlexP)
09:56:40 Quit AlexP (Ping timeout: 240 seconds)
09:56:58 Quit saratoga (Quit: Page closed)
10:11:01 Join AlexP [0] (~alex@rockbox/staff/AlexP)
10:16:57 Quit AlexP (Ping timeout: 240 seconds)
10:21:42 Join AlexP [0] (~alex@rockbox/staff/AlexP)
10:27:07 Quit AlexP (Ping timeout: 240 seconds)
10:28:25 Join AlexP [0] (~alex@rockbox/staff/AlexP)
10:35:07 Quit jhMikeS (Ping timeout: 264 seconds)
10:45:08 Quit AlexP (Ping timeout: 255 seconds)
10:45:28***Saving seen data "./dancer.seen"
10:46:35 Join AlexP [0] (~alex@rockbox/staff/AlexP)
10:58:38 Join wodz [0] (
11:21:00 Quit Topy44 (Ping timeout: 252 seconds)
11:45:33 Quit kugel (Ping timeout: 240 seconds)
11:48:25 Join bcobco [0] (~bcobco@
12:08:02 Quit ender| (Quit: Gray is a color; grey is a colour.)
12:14:48 Quit FOAD (Remote host closed the connection)
12:44:47 Quit Jinx (Ping timeout: 260 seconds)
12:45:31***Saving seen data "./dancer.seen"
12:50:40 Quit ygrek (Ping timeout: 240 seconds)
12:52:08 Join ygrek [0] (~user@
12:56:23 Quit copper (Quit: ZNC -
12:57:29 Join copper [0] (~copper@unaffiliated/copper)
13:05:24 Join ZincAlloy [0] (
13:08:09 Join ygrek_ [0] (~user@
13:10:12 Quit ygrek (Ping timeout: 248 seconds)
13:15:10 Join ikeboy [0] (
13:25:46 Quit Scall (Ping timeout: 264 seconds)
13:34:37 Join Scall [0] (~chat@unaffiliated/scall)
13:39:47kugel_bluebrother: ping
13:40:10 Quit ikeboy (Quit: Leaving)
14:01:19 Quit Marex (Ping timeout: 252 seconds)
14:02:19 Join Marex [0] (~Marex@
14:04:49 Quit bcobco (Remote host closed the connection)
14:05:16 Join bcobco [0] (~bcobco@
14:10:17 Quit ygrek_ (Ping timeout: 240 seconds)
14:22:46 Quit Marex (Ping timeout: 264 seconds)
14:23:29 Join Marex [0] (~Marex@
14:42:23 Join ygrek_ [0] (~user@
14:45:31 Join amayer [0] (
14:45:33***Saving seen data "./dancer.seen"
14:46:30 Join Topy44 [0] (
14:59:20 Quit bcobco (Remote host closed the connection)
14:59:46 Join bcobco [0] (~bcobco@
15:32:02 Quit Strife89 (Ping timeout: 256 seconds)
15:54:43 Quit goom (Ping timeout: 240 seconds)
15:55:45 Nick KotH is now known as shirt (
15:55:50 Nick shirt is now known as KotH (
16:01:42 Quit wodz (Quit: Leaving)
16:04:57 Quit the-kyle (Ping timeout: 240 seconds)
16:05:17 Join kugel [0] (~kugel@
16:05:18 Quit kugel (Changing host)
16:05:18 Join kugel [0] (~kugel@rockbox/developer/kugel)
16:07:18 Join the-kyle [0] (
16:09:55 Quit kugel_ (Ping timeout: 240 seconds)
16:10:10 Quit kugel (Ping timeout: 240 seconds)
16:18:10 Join Glorfindel [0] (b80969c8@gateway/web/freenode/ip.
16:26:09 Quit Glorfindel (Ping timeout: 246 seconds)
16:28:02 Join Strife89 [0] (~Strife89@2602:306:bce1:8c20:8cfa:d1bb:8a33:ec3d)
16:44:13 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
16:44:17 Quit petur (Quit: *plop*)
16:45:37***Saving seen data "./dancer.seen"
16:49:11 Join Glorfindel [0] (~Thunderbi@unaffiliated/glorfindel)
17:10:35GlorfindelWhat are the key controls to shoot on doom on a nano 1st gen?
17:11:48[Saint], iirc.
17:13:32Glorfindelhmm, not working then.
17:15:31Glorfindelit refuses to shoot
17:21:43GlorfindelAnd when I go to the "set keys" section there is no "play" button as an option to set "shoot" to. There are, however, some keys that are not on the player its self
17:23:03[Saint]are you using the Freedoom WAD, or, your own, or what?
17:23:09[Saint]I'll give it a go.
17:24:07[Saint]One sec.
17:27:15[Saint]Working fine here.
17:28:04[Saint]Just as ridiculous, and hard to play as I remember.
17:28:08[Saint]But, working fine.
17:28:25Glorfindelthe shareware version has the same problem
17:28:54[Saint]WHat version of Rockbox are you running?
17:29:29[Saint]Two things to try: 1 - update; 2 - wipe your config (back it up first if you want)
17:29:38[Saint]The releases are ancient.
17:29:48[Saint]I highly advise updating.
17:29:52 Quit us`0gb (Quit:
17:30:18Glorfindelhow would I do that?
17:30:23[Saint]I can't say that it will fix your issue, but I can't say its any good to be running an 18 month old build either.
17:30:48Glorfindeluse the installer?
17:31:02[Saint]You can use or manually via
17:31:56[Saint]I can't say if it will fix this issue with Doom, but you should update anyway, because a lot has changed in the past ~18 months.
17:41:27 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:44:59 Quit AlexP (Ping timeout: 255 seconds)
17:56:37Glorfindelthanks :)
18:04:04 Join maruk1 [0] (
18:07:33 Quit maruk1 (Client Quit)
18:24:57 Join ikeboy [0] (
18:26:12 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:45:41***Saving seen data "./dancer.seen"
18:51:45 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:52:03bluebrotherkugel: pong
18:53:46 Quit ygrek_ (Ping timeout: 256 seconds)
18:54:40 Join Jinx [0] (Dojo@unaffiliated/jinx)
18:55:29 Quit nyanpasu (Quit: brb reboot)
18:56:00kugelbluebrother: I stumbled upon "unifdef", a program to remove #ifdef'ed sections from source code. in contrast to cpp it removes empty lines as well
18:56:45kugeli think we could use that for preprocessing the manual which we always (at least a few years ago) wanted instead of all those opts
18:57:04bluebrothersounds interesting
18:57:26 Join fs-bluebot [0] (
19:01:22kugelit can also handle stuff like defined(X) || defined(Y), works pretty well in my experience
19:02:55bluebrotherI was just looking at its documentation. Sounds like it would be sufficient for our goals. Nice :)
19:05:42 Join us`0gb [0] (
19:14:01 Join lebellium [0] (
19:15:57 Join guymann [0] (~c@unaffiliated/guymann)
19:22:36 Join ygrek_ [0] (~user@
19:31:54 Join pretty_function [0] (~sigBART@
19:40:15lebelliumpamaury (logs): 13:25 + 10:30 = 24 hrs playback time :)
19:41:57 Quit mortalis (Ping timeout: 264 seconds)
19:45:57 Quit ikeboy (Ping timeout: 240 seconds)
19:47:27 Join ikeboy [0] (
19:48:37 Join ender| [0] (krneki@2a01:260:4094:1:42:42:42:42)
20:04:39 Quit Strife89 (Quit: Leaving)
20:09:24 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:17:09 Quit ikeboy (Quit: Leaving)
20:24:19 Quit Glorfindel (Ping timeout: 248 seconds)
20:26:37 Join Glorfindel [0] (~Thunderbi@unaffiliated/glorfindel)
20:35:31 Quit Glorfindel (Ping timeout: 248 seconds)
20:39:13 Quit ender| (Quit: Never in the history of opinions have I been wrong, except for that one time when I thought I was wrong. -- Glen Cook: Wicked Bronze Ambition)
20:40:13 Join ender| [0] (krneki@2a01:260:4094:1:42:42:42:42)
20:45:44***Saving seen data "./dancer.seen"
20:49:02 Quit ygrek_ (Ping timeout: 255 seconds)
21:02:32 Quit amayer (Quit: Leaving)
21:21:33 Quit krnlyng (Read error: Connection timed out)
21:22:34 Join krnlyng [0] (~liar@
21:39:36 Quit pretty_function (Remote host closed the connection)
21:47:59 Quit us`0gb (Ping timeout: 255 seconds)
21:57:36 Quit kugel (Ping timeout: 240 seconds)
22:03:04 Quit Xerion (Read error: Connection reset by peer)
22:03:04 Join Xerion_ [0] (
22:03:14 Nick Xerion_ is now known as Xerion (
22:06:14 Join Sandisk [0] (5f879ef6@gateway/web/freenode/ip.
22:07:26 Quit Sandisk (Client Quit)
22:11:25 Join us`0gb [0] (
22:19:10 Quit Xerion (Read error: Connection reset by peer)
22:19:37 Join Xerion [0] (
22:19:46 Quit Xerion (Remote host closed the connection)
22:30:37 Join Xerion [0] (
22:43:31 Quit us`0gb (Ping timeout: 240 seconds)
22:45:48***Saving seen data "./dancer.seen"
22:55:01lebelliumpamaury: ping
22:56:55 Join us`0gb [0] (
23:00:45 Join RiD [0] (~RiD@
23:02:23pamaurylebellium: pong
23:03:14lebelliumpamaury: did you read logs or were you too busy watching Brazil collapsing? :)
23:16:41 Quit pamaury (Ping timeout: 240 seconds)
23:17:05 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:55:13 Quit pamaury (Ping timeout: 256 seconds)

Previous day | Next day