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 2015-12-17

00:01:25 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
00:55:00 Quit bertrik (Remote host closed the connection)
00:55:12 Quit ender` (Quit: ☀☁☂☃☄★☆☇☈☉☊☋☌☍☎☏☐☑☒☓☚☛☜☝☞☟☠☡☢☣☤☥☦☧☨☩☪☫☬☭☮☯☰☱☲☳☴☵☶☷☸☹☺☻☼☽☾☿♀♁♂♃♄♅♆♇♈♉♊♋♌♍♎♏♐♑♒♓♔♕♖♗♘♙)
00:55:35 Join pablo_pi_ [0] (~pablo@
00:56:34 Quit __builtin (Read error: Connection reset by peer)
00:57:18 Join [Franklin] [0] (~me@unaffiliated/franklin)
00:57:52 Quit pablo_pi (Ping timeout: 272 seconds)
00:59:35 Quit FSanches (Quit: Leaving.)
01:03:39 Quit krnlyng (Ping timeout: 265 seconds)
01:11:15 Join alexweissman [0] (
01:14:56***Saving seen data "./dancer.seen"
01:16:18 Join krnlyng [0] (
01:28:54 Nick [Franklin] is now known as __builtin (~me@unaffiliated/franklin)
01:31:59 Quit __builtin (Quit: ?f u cn rd ths, u cn gt a gd jb n sftwr tstng.? ? Anonymous)
01:42:52 Quit JdGordon (Ping timeout: 255 seconds)
01:48:45 Join JdGordon [0] (
01:48:45 Quit JdGordon (Changing host)
01:48:45 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
01:53:24 Quit JdGordon (Ping timeout: 240 seconds)
01:53:45 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
01:58:16 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
01:58:49 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
01:59:38 Quit JdGordon (Ping timeout: 260 seconds)
02:11:10 Quit pamaury (Ping timeout: 256 seconds)
02:13:57 Quit Bray90820 ()
02:25:56 Join JdGordon [0] (
02:25:56 Quit JdGordon (Changing host)
02:25:56 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
02:29:09 Join Bray90820 [0] (
02:29:18 Quit JdGordon_ (Ping timeout: 256 seconds)
02:31:00 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
02:31:04 Quit JdGordon (Ping timeout: 240 seconds)
02:32:19 Quit Bray90820 (Read error: Connection reset by peer)
02:32:52 Join Bray90820 [0] (
02:32:52 Quit Bray90820 (Client Quit)
02:33:11 Join PurlingNayuki [0] (~Thunderbi@
02:33:24 Join Bray90820 [0] (
02:34:21 Quit ZincAlloy (Quit: Leaving.)
02:38:21 Join JdGordon [0] (
02:38:21 Quit JdGordon (Changing host)
02:38:21 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
02:40:33 Quit JdGordon_ (Ping timeout: 240 seconds)
02:53:33 Quit alexweissman (Ping timeout: 240 seconds)
02:59:54 Quit puckipedia (Ping timeout: 256 seconds)
03:01:41 Join puckipedia [0] (
03:03:53 Quit krnlyng (Ping timeout: 240 seconds)
03:07:58 Join chrisb [0] (
03:10:30 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
03:10:32 Quit Poodlemastah (Ping timeout: 250 seconds)
03:13:40 Quit JdGordon (Ping timeout: 265 seconds)
03:14:57***Saving seen data "./dancer.seen"
03:16:01 Join Poodlemastah [0] (
03:17:00 Join krnlyng [0] (
03:32:26 Join [Franklin] [0] (~me@unaffiliated/franklin)
03:52:36 Quit Bray90820 (Ping timeout: 256 seconds)
03:55:00 Join Bray90820 [0] (
04:00:21 Join alexweissman [0] (
04:05:22 Quit alexweissman (Ping timeout: 272 seconds)
04:22:45 Nick [Franklin] is now known as __builtin (~me@unaffiliated/franklin)
04:38:36 Join fs-bluebot_ [0] (
04:38:42 Quit bluebrother (Disconnected by services)
04:38:46 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
04:40:25 Quit fs-bluebot (Ping timeout: 240 seconds)
04:43:36 Join nat__ [0] (3fe3d2ee@gateway/web/freenode/ip.
04:43:43 Quit nat__ (Client Quit)
04:52:41 Quit krnlyng (Ping timeout: 256 seconds)
05:02:11 Quit chrisb (Ping timeout: 246 seconds)
05:06:20 Quit TheSeven (Ping timeout: 260 seconds)
05:06:29 Join krnlyng [0] (
05:07:35 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:15:01***Saving seen data "./dancer.seen"
05:15:38 Join krabador [0] (~krabador@unaffiliated/krabador)
05:18:35 Quit JanC (Ping timeout: 240 seconds)
05:24:48 Join Bray90820_ [0] (~bray90820@2604:2d80:8033:8583:dd48:9f2a:e6b3:8c9e)
05:24:56 Quit Bray90820 (Ping timeout: 246 seconds)
05:28:24 Join Bray90820 [0] (
05:30:15 Quit Bray90820_ (Ping timeout: 260 seconds)
05:32:19 Join JanC [0] (~janc@lugwv/member/JanC)
06:03:30 Quit TheSeven (Ping timeout: 260 seconds)
06:04:46 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:32:14 Quit krabador (Ping timeout: 255 seconds)
06:32:52 Quit ungali (Remote host closed the connection)
06:33:56 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
06:34:01 Join ungali [0] (
06:34:02 Quit ungali (Changing host)
06:34:02 Join ungali [0] (ungali@unaffiliated/ungali)
06:49:17 Join alexweissman [0] (
06:53:05 Quit krnlyng (Ping timeout: 260 seconds)
06:55:21 Quit [Saint] (Remote host closed the connection)
07:00:50 Join [Saint] [0] (~hayden@rockbox/staff/saint)
07:05:54 Join krnlyng [0] (~liar@
07:06:10 Quit pamaury (Ping timeout: 265 seconds)
07:15:04***Saving seen data "./dancer.seen"
07:18:19 Quit sparetire_ (Quit: sparetire_)
07:30:44 Quit Bray90820 (Ping timeout: 255 seconds)
07:57:06 Join Bray90820_ [0] (~bray90820@2002:ad11:2e75:0:9d19:c85f:e56d:7a55)
08:22:11 Join ender` [0] (
08:39:36 Join xorly [0] (
08:50:21 Quit krnlyng (Ping timeout: 272 seconds)
09:03:43 Join krnlyng [0] (
09:15:06***Saving seen data "./dancer.seen"
09:18:19 Join wodz [0] (
09:28:13fs-bluebot_Build Server message: New build round started. Revision f753b8e, 255 builds, 25 clients.
09:35:12fs-bluebot_Build Server message: Build round completed after 420 seconds.
09:35:13fs-bluebot_Build Server message: Revision f753b8e result: 0 errors 19 warnings
09:35:14fs-bluebot_Build Server message: New build round started. Revision 5c67ebb, 255 builds, 25 clients.
09:41:05fs-bluebot_Build Server message: Build round completed after 352 seconds.
09:41:06fs-bluebot_Build Server message: Revision 5c67ebb result: 0 errors 11 warnings
09:44:53 Join petur [0] (~petur@rockbox/developer/petur)
10:02:28 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:10:53 Join edhelas [0] (535592e7@gateway/web/freenode/ip.
10:10:58edhelashi everyone
10:12:48edhelasI have a small issue with Rockbox on my iPod Mini that uses a CF to SDCard converter
10:14:59edhelaswhen having a playlist, the playback just stop at the last second of each FLAC song (with an horrible "sound loop" bug)
10:16:39 Quit pamaury (Ping timeout: 260 seconds)
10:25:15 Quit PurlingNayuki (Quit: PurlingNayuki)
10:55:23fs-bluebot_Build Server message: New build round started. Revision 2d23c03, 255 builds, 27 clients.
11:01:04fs-bluebot_Build Server message: Build round completed after 341 seconds.
11:01:05fs-bluebot_Build Server message: Revision 2d23c03 result: 0 errors 24 warnings
11:06:52 Quit krnlyng (Ping timeout: 265 seconds)
11:15:07***Saving seen data "./dancer.seen"
11:15:26 Quit girafe (Read error: Connection reset by peer)
11:16:23 Quit megal0maniac (Killed ( (Nickname regained by services)))
11:16:25 Join megal0maniac [0] (~megal0man@unaffiliated/megal0maniac)
11:19:27 Join krnlyng [0] (
11:49:15 Join krabador [0] (~krabador@unaffiliated/krabador)
12:04:23 Quit xorly (Ping timeout: 265 seconds)
12:26:51 Quit krabador (Quit: Sto andando via)
12:37:30 Join TheLemonMan [0] (~lemonboy@unaffiliated/thelemonman)
12:39:17 Quit gartral (Read error: Connection reset by peer)
13:07:19 Quit krnlyng (Ping timeout: 255 seconds)
13:15:08***Saving seen data "./dancer.seen"
13:20:05 Join krnlyng [0] (
13:30:44wodzprof_wolfff: setting charging for 500mA by default is not standard compliant. You are allowed to do so a) after enumeration and exchanging current draw information b) when you know it is wall charger (particular pull-ups on data lines IIRC)
13:31:40gevaertsOr (c) if you decide that everyone does it so one more won't change much :)
13:41:32gevaertsI prefer handling it properly, true
13:44:05prof_wolfffwodz: really it was changed to be 'in tune' with the bootloader that charges battery using that limit, and sometimes it is also needed because the HDD is reading for a while (big database) before RB raises the limit to 500mA, if battery is low it drains fast and then the HDD hangs and reboots forever
13:44:53wodzprof_wolfff: Does it help current usb driver to use uncached mem? I thought usb on classic is broken?
13:46:10prof_wolfffi am not so sure, some people says that latest versions do not disconnect USB, i guess if it is due the 500mA limit, it always was 1000mA
13:47:20prof_wolfffi am refering to this patch already commited:
13:48:09*wodz encountered parse error :-(
13:48:16gevaertsIt drew 1000mA without asking? Yes, on some hardware that will get you disconnected...
13:48:29wodzah, that way
13:48:53gevaertsThe same could happen with 500mA, but I expect the current limiters (that I believe already aren't *that* common) won't be configurable
13:49:01wodzI am surprised *it worked* on some
13:49:13edhelassomeone has an idea about my little problem ? I suspect some hardware related issue but I'm not sure
13:49:37prof_wolfffit happened for me, and this patch solves all my issues, iPod can drain more than 700mA when battery is charging and HDD reading/writting
13:50:15prof_wolfffproblem is the voltage drop then current increases above 500mA
13:50:34prof_wolfff*when the current*
13:50:57gevaertsedhelas: I've never heard of that particular issue. It could be hardware I guess, but I really have no idea. What happens if you try playing from a directory with more than one file?
13:51:12gevaertsAh, right, voltage drop sounds more likely
13:51:13 Join krabador [0] (~krabador@unaffiliated/krabador)
13:52:26prof_wolfffit should be the USB hub, when surge current is above 'specifications' then voltage decreases to compensate available power (i guess)
13:53:50gevaertsThat sounds very plausible to me at least
13:53:58edhelasgevaerts: same issue, the playback just stop at the last second and loop on a weird noise
13:54:26gevaertsedhelas: have you tried checking the filesystem for corruption?
13:55:26edhelasI already format the whole thing with iTunes and put Rockbox back but still, it's weird that it works perfectly with OGG files and have this issue on FLAC files only
13:55:49edhelasalso I didn't try with other formats (I only have OGG and FLAC)
13:56:19prof_wolfffgevaerts: it is also compatible with people experimenting USB disconnects 'some days and not other days' because disconnects are more likely when battery is low
13:56:48gevaertsprof_wolfff: yes, indeed. It also fits with the hard to reproduce pattern
13:57:09gevaertsedhelas: I assume those flac files play fine elsewhere?
13:57:38edhelasyes, I'll double check the MD5 to be sure
13:58:15prof_wolfffgevaerts: battery charges at a maximum of ~330mA when it is empty, then current decreases to ~0mA as it charges
13:58:44gevaertsedhelas: you could try with the rockbox simulator (from"> if you're on windows) to *really* rule out hardware
13:59:27gevaerts(unpack, put your audio files in the simdisk/ directory, launch the exe)
13:59:50edhelasmhhh don't have windows for now, but I can try later
14:00:04gevaertsmac? linux?
14:00:16edhelasosx and debian
14:01:15gevaertsok. We don't have builds for either available, but I might be able to do a debian build for you. Which debian version? 64 bit?
14:01:41edhelasyup, jessie
14:04:15*gevaerts finds a jessie vm
14:05:10edhelasok, the md5sum is different from the files
14:06:22gevaertsOK, file corruption. Could be a dodgy sd adapter...
14:07:38edhelasor the tool that I'm using to copy the files, let me check :)
14:08:11gevaertsOr you forgot you edited tags or something like that
14:09:17edhelaswell I'm using Cantata to transfer, and it maybe change some tags and rename, thanks for your time <3
14:09:32edhelasI'll maybe submit a new theme for iPod Mini also
14:09:56gevaertsOf course changed tags shouldn't have the effect you see
14:12:59edhelasahah you're right, with a raw transfer of the file it seems fine :) so I'll try to see with the Cantata developper why its device transfer tool create corrupted FLAC files
14:18:04gevaertsOK, so you probably won't need this simulator after all :)
14:19:09edhelasno thanks :) by the way do you know if some already tried to use EyeFi SD card with Rockbox ?
14:19:50gevaertsNot that I know of. I think people find them too expensive to buy on the off-chance that they might work :)
14:21:12edhelasyeah I can understand that
14:21:56edhelaswell you made a rockbox user happy today :D
14:28:32 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
14:30:24 Quit JdGordon_ (Ping timeout: 260 seconds)
14:32:05 Quit krnlyng (Ping timeout: 240 seconds)
15:04:50 Join krnlyng [0] (
15:14:13 Quit JdGordon (Ping timeout: 255 seconds)
15:15:09***Saving seen data "./dancer.seen"
15:17:21 Join JdGordon [0] (~jonno@
15:17:21 Quit JdGordon (Changing host)
15:17:21 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
15:29:16 Quit ungali (Remote host closed the connection)
15:38:30 Quit wodz (Quit: Leaving)
15:47:01 Quit maraz (Ping timeout: 260 seconds)
16:23:42 Join xorly [0] (
16:24:30 Join PurlingNayuki [0] (~Thunderbi@
16:27:19 Quit krnlyng (Ping timeout: 265 seconds)
16:32:35 Join maraz [0] (maraz@
16:32:54 Quit PurlingNayuki (Ping timeout: 240 seconds)
16:38:09 Quit xorly (Ping timeout: 260 seconds)
16:42:44 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
16:45:59 Quit JdGordon (Ping timeout: 272 seconds)
16:59:29 Join einhirn [0] (~Miranda@
17:00:47 Quit krabador (Quit: Sto andando via)
17:02:42 Join girafe [0] (
17:05:31 Join krnlyng [0] (
17:15:10***Saving seen data "./dancer.seen"
17:28:07 Join FSanches [0] (~felipe@2804:14c:37:268b:6429:f046:1417:50c8)
17:28:09 Join JdGordon [0] (
17:28:09 Quit JdGordon (Changing host)
17:28:09 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
17:30:54 Quit JdGordon_ (Ping timeout: 240 seconds)
17:35:52 Join sparetire_ [0] (~sparetire@unaffiliated/sparetire)
17:40:11 Quit edhelas (Ping timeout: 252 seconds)
17:41:15 Quit petur (Quit: *plop*)
17:49:18 Join Electricguy [0] (
18:37:57 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
18:41:09 Quit JdGordon (Ping timeout: 256 seconds)
18:42:52 Join JdGordon [0] (
18:42:52 Quit JdGordon (Changing host)
18:42:52 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
18:44:06 Quit JdGordon_ (Ping timeout: 265 seconds)
18:53:24 Quit shamus (Ping timeout: 240 seconds)
18:54:20 Join shamus [0] (
19:00:45 Join krnlyng_ [0] (~liar@
19:00:47 Join ZincAlloy [0] (
19:01:52 Nick suYin`OFF is now known as suYin (
19:02:28 Quit krnlyng (Ping timeout: 265 seconds)
19:05:31 Quit krnlyng_ (Ping timeout: 256 seconds)
19:12:41 Join krnlyng [0] (~liar@
19:14:14 Quit FSanches (Quit: Leaving.)
19:14:31 Join lebellium [0] (
19:15:14***Saving seen data "./dancer.seen"
19:24:09 Quit girafe (Read error: Connection reset by peer)
19:38:00 Join Strife89|M105 [0] (~Strife89@2602:306:bce1:8c20:9c63:a7af:f680:8aa0)
19:38:08 Join Link8 [0] (
19:40:25 Quit Link8 (Remote host closed the connection)
20:03:01 Quit Smx (Excess Flood)
20:05:00 Join Smx [0] (Elite8556@gateway/shell/elitebnc/x-vbbyatfcvvzqmlqr)
20:11:46 Join krabador [0] (~krabador@unaffiliated/krabador)
20:20:16 Join FSanches [0] (~felipe@2804:14c:37:268b:6429:f046:1417:50c8)
20:27:11 Join girafe [0] (
20:28:20 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
20:30:49 Quit krnlyng (Ping timeout: 272 seconds)
20:44:06 Join krnlyng [0] (
21:06:33 Quit FSanches (Quit: Leaving.)
21:14:20 Quit alexweissman (Remote host closed the connection)
21:15:16***Saving seen data "./dancer.seen"
21:33:28 Join FSanches [0] (~felipe@2804:14c:37:268b:6429:f046:1417:50c8)
21:59:12 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
22:01:34 Quit JdGordon (Ping timeout: 240 seconds)
22:03:49 Quit JdGordon_ (Ping timeout: 246 seconds)
22:04:08 Join JdGordon [0] (
22:04:08 Quit JdGordon (Changing host)
22:04:08 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
22:14:56 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
22:19:02 Quit FSanches (Quit: Leaving.)
22:24:56 Join alexweissman [0] (
22:29:04 Quit alexweissman (Ping timeout: 240 seconds)
22:54:37 Join alexweissman [0] (
22:55:12 Join dfkt [0] (~dfkt@unaffiliated/dfkt)
22:59:04 Quit Electricguy (Ping timeout: 255 seconds)
23:04:48 Quit TheLemonMan (Quit: "It's now safe to turn off your computer.")
23:05:14 Quit JanC (Ping timeout: 240 seconds)
23:06:27[Saint]Is there an easy way of submitting all the committed changes in a given branch to gerrit?
23:06:49[Saint]Or do I seriously have to reverse through my tree and submit each one-by-one (please say no...)
23:07:54 Quit rela (Ping timeout: 240 seconds)
23:08:23 Quit lebellium (Quit: ChatZilla 0.9.92 [Firefox 43.0/20151208100201])
23:10:33[Saint]There's a couple of changes I need to discuss as well.
23:10:48[Saint]The main one that is bothering me is this:
23:11:46[Saint]Should I bother jumping down the rabbit hole of detecting the currently installed Android NDk version, or just say "Nope, fuck you, use the latest one" and patch the code that deals with it accordingly as required?
23:12:36[Saint](our threading seems to be pissing off the gold linker - a lot of projects are having this problem with threading and the gold linker in NDK r10e
23:13:55[Saint]Since changing the way we handle threading is an incredibly non-trivial change, the obvious solution seems to me to just stop using gold and rely on bfd - which seems to have no issue with our threading at all in this regard.
23:15:20***Saving seen data "./dancer.seen"
23:19:16*[Saint] is also trying to work out a sane way of handling static allocation for our main buffer on Android
23:19:33[Saint]At the moment, just giving everything 8MB is:
23:19:37[Saint]A - Stupid
23:19:47 Join JanC [0] (~janc@lugwv/member/JanC)
23:19:49[Saint]B - Broken
23:21:27[Saint]Instead of checking for specific resolutions and orientations (no idea why the original patch for this cares about orientation), I'm thinking it makes a lot more sense to just define a few "buckets" for static allocation to fall into.
23:23:21[Saint]if LCD_WIDTH * LCD_HEIGHT >= N; memory=X | else memory=Y | else memory=z...etc.
23:23:38 Nick suYin is now known as suYin`OFF (
23:37:34 Quit dfkt (Read error: Connection reset by peer)
23:38:39[Saint]gevaerts: I know you were (rightly) skeptical of the original patch to tackle this which lies abandoned at g#956
23:38:39[Saint]Instead of looking at LCDWIDTH and LCDHEIGHT directly, and bothering to care about orientation (I have no idea why the original author did this), is forgetting about orientation entirely and looking at total pixels (LCDWIDTH * LCDHEIGHT) a more acceptable solution for this?
23:38:45[Saint]...something needs to be done here.
23:38:51fs-bluebot_Gerrit review #956 at : increase memorysize for Android build. by Chiwen Chang
23:41:04[Saint]The way I was looking at calculating the static allocation of the buffer was to do 8MB + LCDWIDTH * LCDHEIGHT * 2, which gives room for the two fullscreen bitmaps that the theme engine will want to allocate at the very least.
23:42:42[Saint]The size of the remaining audiobuffer isn't particularly important on these devices where the SoC is usually quite powerful and capable, can deep sleep, and the storage is quite fast.
23:43:33[Saint]The need for a (semi-) dynamic buffer allocation comes almost entirely from the theme engine.
23:44:07[Saint]Once you get up past 540p resolution almost all of the buffer allocation goes to the theme engine.
23:44:15[Saint]Up past 720p, all of it does.
23:45:05gevaerts[Saint]: I don't seem to remember that!
23:45:37[Saint]In that case - thoughts on it now? ;)
23:46:20[Saint]The implementation in the linked patch is indeed fairly stupid - but I'm wondering if mine is any less stupid, or if you have another idea that you might find acceptable.
23:46:54 Quit Strife89|M105 (Ping timeout: 240 seconds)
23:47:06[Saint]Mine is very similar but it removes the need to care about specific resolutions and orientations and only looks at the total amount of pixels.
23:47:15gevaertsYes, that's what I'd do too
23:48:23gevaertsI have no opinion on how much you need, but something like constant + pixels * anotherconstant seems sensible
23:48:39gevaertsPossibly round up to the nearest deliberate-looking number :)
23:49:04*[Saint] nods
23:49:51*gevaerts knows that one needs powers of two and possibly not too complicated multiples of powers of two in order to be taken seriously
23:50:42gevaertsYou can have 8MB, or 24MB, or maybe even 56MB, but *not* 27.3MB!
23:51:17[Saint]I know that the theme engine wants at least two fullscreen bitmaps, so I used that as the basis for the minimum required buffer calculation. 8MB (default for all the app targets) + LCDHEIGHT * LCDWIDTH * 2 seems to work out to a roughly sensible number.
23:51:28[Saint]Though there's no sane way to take fonts into account.
23:51:48[Saint]For the very large screen devices, a single font could quite easily absorb all that buffer and more.
23:53:32[Saint]IIRC, my antialiased 50pt GNU Unifont tops out at about 80MB, and by default we want to keep 200 glyphs in the cache from memory. That's around 6~8MB or so in itself.
23:53:57[Saint]Static allocation is a bit of a bitch in this case.
23:54:20gevaertsYes, as long as we allocate one static chunk we'll keep having trouble
23:54:26[Saint]It makes a lot more sense for it to be dynamically allocated, but that's a really radical change for the hosted targets.
23:54:41gevaertsBut then that's what a standard java vm does, so it must be right :)
23:54:51[Saint]Ha! :)

Previous day | Next day