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 2013-04-28

00:03:58 Quit stoffel (Remote host closed the connection)
00:04:34 Join stoffel [0] (~quassel@pD9E43D6F.dip0.t-ipconnect.de)
00:07:53 Quit stoffel (Remote host closed the connection)
00:18:13 Quit Misanthropos (Quit: Verlassend)
00:20:42 Join lebellium_ [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
00:21:22 Quit lebellium (Ping timeout: 245 seconds)
00:21:22 Nick lebellium_ is now known as lebellium (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
00:27:47 Join habys [0] (~luke@arikui.org)
00:40:34 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
00:48:06 Quit [7] (Remote host closed the connection)
00:51:58 Quit y4n (Quit: AMIGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHAHAHAHAAAAAAAAAAAAHAHAAA)
00:53:57 Quit kaitsu1 (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com ))
01:00
01:04:47 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
01:07:36***Saving seen data "./dancer.seen"
01:08:18 Quit ur725 (Quit: Leaving)
01:10:06 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
01:32:22 Quit lebellium (Quit: ChatZilla 0.9.90 [Firefox 21.0/20130423212553])
01:51:29 Quit prof_wolfff (Ping timeout: 258 seconds)
01:51:30 Quit dfkt_ (Remote host closed the connection)
01:52:50 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
01:52:52 Quit bertrik_ (Read error: Connection reset by peer)
01:57:04 Quit bertrik (Ping timeout: 245 seconds)
01:58:17 Quit kiwicam (Remote host closed the connection)
01:58:47 Join kiwicam [0] (~quassel@101.98.163.139)
02:00
02:47:50 Join jhMikeS [0] (~jethead71@50.4.247.132)
02:47:50 Quit jhMikeS (Changing host)
02:47:50 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
03:00
03:07:39***Saving seen data "./dancer.seen"
03:21:28 Quit bluebrother (Disconnected by services)
03:21:33 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
03:22:18 Quit Raptors (Quit: Leaving)
03:23:29 Quit fs-bluebot (Ping timeout: 258 seconds)
03:24:52 Join fs-bluebot [0] (~fs-bluebo@f053152042.adsl.alicedsl.de)
03:38:58 Nick [Saint_] is now known as [Saint] (~saint@rockbox/user/saint)
03:50:30 Join Raptors [0] (~whoneedsa@216-58-33-203.cpe.distributel.net)
03:59:10 Quit derf (Remote host closed the connection)
04:00
04:33:13 Join dhrasmus [0] (~dhrasmus@173-23-133-55.client.mchsi.com)
04:33:36 Quit krabador (Quit: Sto andando via)
04:39:46 Quit pixelma (Read error: Operation timed out)
04:39:46 Quit amiconn (Disconnected by services)
04:39:48 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:39:52 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:40:42 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
05:00
05:07:40***Saving seen data "./dancer.seen"
05:07:57 Join TheSphinX_ [0] (~briehl@p57A38293.dip0.t-ipconnect.de)
05:11:28 Quit TheSphinX^ (Ping timeout: 252 seconds)
05:16:04 Nick c0decat is now known as mastertux (~mrtux@unaffiliated/mrtux)
05:20:59 Nick mastertux is now known as mrtux (~mrtux@unaffiliated/mrtux)
05:29:38 Quit dhrasmus (Quit: Leaving)
05:40:14 Quit TheSeven (Disconnected by services)
05:40:23 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
07:00
07:07:43***Saving seen data "./dancer.seen"
07:16:52 Join |akaWolf| [0] (~akaWolf@188.134.9.161)
07:17:29 Quit SuperBrainAK (Quit: pbly gone to sleep (-.-)Zzz...)
07:20:13 Join derf [0] (~derf@fuzzyneural.net)
07:37:38 Join n1s [0] (~n1s@nl118-168-30.student.uu.se)
07:37:39 Quit n1s (Changing host)
07:37:39 Join n1s [0] (~n1s@rockbox/developer/n1s)
08:00
08:01:00 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
08:02:35 Quit |akaWolf| (Ping timeout: 264 seconds)
08:14:47 Quit kevku (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/)
08:18:23 Quit jhMikeS (Read error: Connection reset by peer)
08:52:36 Join kaitsu1 [0] (~bubba@a91-152-170-133.elisa-laajakaista.fi)
09:00
09:04:54 Join melmothX [0] (~melmoth@unaffiliated/melmothx)
09:07:46***Saving seen data "./dancer.seen"
09:08:33 Quit melmothX (Client Quit)
09:17:08 Join melmothX [0] (~melmoth@unaffiliated/melmothx)
09:27:47 Quit froggyman (Ping timeout: 264 seconds)
09:45:12 Join webguest55 [0] (~59f2cd53@www.haxx.se)
09:45:12 Quit webguest55 (Client Quit)
10:00
10:12:11 Join stoffel [0] (~quassel@pD9E4300B.dip0.t-ipconnect.de)
10:14:59 Join jhMikeS [0] (~jethead71@50.4.247.132)
10:14:59 Quit jhMikeS (Changing host)
10:14:59 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
10:51:59 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:00
11:07:34 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
11:07:50***Saving seen data "./dancer.seen"
11:08:08 Join lebellium [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
11:17:10 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
11:33:11 Quit [Saint] (Remote host closed the connection)
11:34:20 Join [Saint] [0] (~saint@rockbox/user/saint)
11:51:30 Join lorenzo92 [0] (~chatzilla@host23-109-dynamic.0-87-r.retail.telecomitalia.it)
11:59:50 Join prof_wolfff [0] (~prof_wolf@62.83.50.196.dyn.user.ono.com)
12:00
12:00:00 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
12:02:45 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
12:05:59 Join dfkt [0] (dfkt@unaffiliated/dfkt)
12:06:44 Quit bertrik (Ping timeout: 252 seconds)
12:22:37 Join kaputnik__ [0] (~kaputnik@p5DD9EA1E.dip0.t-ipconnect.de)
12:26:32 Quit jhMikeS (Ping timeout: 252 seconds)
12:51:27 Quit kaputnik__ (Ping timeout: 256 seconds)
12:58:17 Quit kevku (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/)
13:00
13:07:54***Saving seen data "./dancer.seen"
13:10:00 Join sakax [0] (~sakax@d8D862D2D.access.telenet.be)
13:29:08 Quit LjL (Ping timeout: 256 seconds)
13:40:13 Quit Lynx_ (Ping timeout: 252 seconds)
13:41:10 Join Lynx_ [0] (~bayer@109.171.130.211)
13:55:17 Quit stoffel (Ping timeout: 264 seconds)
15:00
15:03:31 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
15:07:58***Saving seen data "./dancer.seen"
15:25:57 Quit prof_wolfff (Ping timeout: 245 seconds)
15:51:57bertrik_anyone here with an AMSv2 player and a class 10 sdhc microsd card?
15:52:29bertrik_there's been reports that some class 10 microsd don't work on the clip zip
15:52:54bertrik_I'm not convinced yet if "class 10" has anything to do with it though
16:00
16:03:18 Quit lorenzo92 (Remote host closed the connection)
16:12:53 Nick bertrik_ is now known as bertrik (~quassel@rockbox/developer/bertrik)
16:15:31lebelliumI have a clip Zip with Samsung 32GB class 10 card
16:15:44lebelliumand it works
16:17:02lebelliumbertrik^
16:20:37funmanbertrik: there's been many reports of microsd cards not working on AMS(v*), class 10 or not
16:21:41funmanIMHO the card class has as much importance as the color of the packaging in which the card came ..
16:29:27 Quit pamaury (Ping timeout: 246 seconds)
16:32:43 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
16:32:48 Quit pamaury (Client Quit)
16:33:00 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
16:34:58 Join kaputnik__ [0] (~kaputnik@p5DD9EA1E.dip0.t-ipconnect.de)
16:39:19bertrikfunman: agreed, although class 10 indicates that at least it's not an ancient card
17:00
17:01:39 Quit pamaury (Ping timeout: 246 seconds)
17:08:01***Saving seen data "./dancer.seen"
17:08:33soapThough (was it CHDK?) I've seen benchmarks showing severe compromises are made sometimes to achieve class rating. As in the random write speed may be an order of magnitude (or more!) slower.
17:16:55 Quit lebellium (Quit: ChatZilla 0.9.90 [Firefox 21.0/20130423212553])
17:20:29 Join krabador [0] (~krabador_@unaffiliated/krabador)
17:25:04 Quit kaputnik__ (Ping timeout: 255 seconds)
17:50:00 Quit kevku (Ping timeout: 245 seconds)
18:00
18:02:50 Join mortalis [0] (~mortalis@77.108.98.176)
18:03:26 Join bp0 [0] (~bp@unaffiliated/bp0)
18:12:35*[Saint] grumbles
18:13:22[Saint]Why....why oh why...is the skin engine as far as a user is concerned totally unaware of the device resolution?
18:13:29[Saint]Grrr.
18:20:36 Quit akaWolf (Quit: my exit)
18:23:03 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
18:23:54 Join lebellium [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
18:27:04[Saint]things would be SOOOOOOO much easier for me if I could query the UI/child viewport dimensions...
18:27:38[Saint]Assuming I could, we'd need three cabbies for all of our targets.
18:27:59[Saint](and each theme would be absurdly complicated, but...hey)
18:28:13gevaertsWell
18:28:22gevaertsI'm not sure that would be a good trade-off
18:28:43gevaertsI mean, people will see cabbie first, and attempting to change it a bit is a natural first thing to do
18:30:09[Saint]Yes, this is true. It could be commented sufficiently. I just severely dislike umpteen dozen basically identical cabbies all with messed up color values and slightly different conventions and positionings.
18:31:24gevaertsNow if you could use the C preprocessor to generate cabbies for all players from one source, I'd be all for it
18:31:52[Saint]Knowing the resolution would simply be good for knowing what bitmaps to preload. Alignment could /almost/ handle itself.
18:32:26[Saint]Well, you could use negative values to align /most/ things. Meaning less complication.
18:40:29 Join desowin [0] (~desowin@pandora.barbara.ds.polsl.pl)
18:45:00 Join froggyman [0] (~me@dhcp-155-92-103-232.nebula.msoe.edu)
18:45:00 Quit froggyman (Changing host)
18:45:00 Join froggyman [0] (~me@unaffiliated/froggyman)
18:45:14 Join dv__ [0] (~quassel@chello080108009040.14.11.vie.surfer.at)
18:45:44 Quit dv_ (Ping timeout: 258 seconds)
18:59:53 Join stoffel [0] (~quassel@pD9E424AC.dip0.t-ipconnect.de)
19:00
19:00:25 Quit krabador (Remote host closed the connection)
19:04:32 Join mt [0] (~quassel@196.218.41.112)
19:04:55 Nick mt is now known as Guest50616 (~quassel@196.218.41.112)
19:08:04***Saving seen data "./dancer.seen"
19:10:55 Nick [Saint] is now known as sinner (~saint@rockbox/user/saint)
19:11:10 Nick sinner is now known as [Saint] (~saint@rockbox/user/saint)
19:34:07 Quit sakax (Quit: Leaving)
19:37:45 Quit stoffel (Remote host closed the connection)
19:40:23 Join diusdf [0] (~guest-ttU@a88-114-108-39.elisa-laajakaista.fi)
19:40:52diusdfhello, so it has become time for me to buy a new mp3-player and I have read that sansa clip+ is possible to be rockboxed
19:41:16diusdfI read from some thread that the 8GB version might have some problems with big SD cards but do you think that is still the case?
19:41:43diusdffor example would it be possible for me to acquire a 32gb sd card and use it fine with 8gb sansa clip+ with rockbox firmware?
19:42:34[Saint]Yes.
19:46:09diusdfsmooth
19:46:34diusdfwonder if the cheapest place to get it is amazon
19:49:36 Join prof_wolfff [0] (~prof_wolf@62.83.50.196.dyn.user.ono.com)
19:49:36 Quit shamus (Read error: Connection reset by peer)
19:50:38 Join shamus [0] (~shmaus@ip-206-192-195-49.marylandheights.ip.cablemo.net)
19:54:51 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
19:55:38 Join liar [0] (~liar@clnet-p09-185.ikbnet.co.at)
19:58:23bertrikI ordered another sansa clip zip for myself, my old one has a problem in the headphone connector. Perhaps I can send this old one to another developer?
19:59:35diusdfaint there howto pages in the net how to fix it like really easily, did it break easily?
19:59:43diusdfwould be good to know for the futures sake
20:00
20:00:59bertrikI think it broke at some point when falling off my desk, still attached to the headphones
20:01:30bertrikProbably would be easy to fix if you have the tools to open it without damaging it
20:01:44bluebrother^so ... what's different between loading Rockbox and rolo-ing it?
20:01:58gevaertsbluebrother^: hardware initialisation, presumably
20:02:09bluebrother^FS #12857 appears only when starting Rockbox, not when rolo-ing it.
20:02:10fs-bluebothttp://www.rockbox.org/tracker/task/12857 Data abort error on iPod Mini 2G (bugs, unconfirmed)
20:03:07bertrikrolo-ing skips the bootloader, but the bootloader should do the same initialisation as the main image
20:04:40*[Saint] hands bertik a guitar pick and a 00 phillips driver
20:04:48bluebrother^find_addr.pl points to disk_partinfo
20:06:03bluebrother^hmm. find_addr.pl requires me to install additional Perl modules :/
20:07:16 Quit kaitsu1 (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com ))
20:07:20[Saint]They're called oysters. ;)
20:08:48bluebrother^hmm. disk_partinfo() only returns a pointer to a local array element :o
20:09:06[Saint]nice.
20:09:56bluebrother^but according to the data abort it's the problem?
20:10:05 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:13:21bluebrother^now if I could read that assembler variant ...
20:20:52 Quit Guest50616 (Ping timeout: 276 seconds)
20:26:29 Quit mortalis (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/)
20:31:53 Join Wardo [0] (~Mirandaha@176-120-190-109.dsl.ovh.fr)
20:36:29 Quit diusdf (Quit: Lähdössä)
20:36:31bluebrother^hmm. The final link seems to use −−gc-sections after that change
20:51:05 Join strawberrytea [0] (~5483f1a2@www.haxx.se)
20:55:08 Join kaputnik__ [0] (~kaputnik@port-92-206-79-54.dynamic.qsc.de)
20:58:39 Quit strawberrytea (Quit: CGI:IRC (EOF))
21:00
21:02:35 Quit kevku (Ping timeout: 264 seconds)
21:06:09 Quit prof_wolfff (Ping timeout: 256 seconds)
21:08:05***Saving seen data "./dancer.seen"
21:09:54fs-bluebotBuild Server message: New build round started. Revision 850491a, 214 builds, 39 clients.
21:09:59bluebrother^now it gets interesting ...
21:12:02 Join SuperBrainAK [0] (~Andy@67-42-37-93.phnx.qwest.net)
21:12:17*bluebrother^ crosses fingers
21:15:48 Quit kaputnik__ (Ping timeout: 248 seconds)
21:16:38fs-bluebotBuild Server message: Build round completed after 404 seconds.
21:17:15 Quit akaWolf (Ping timeout: 252 seconds)
21:17:34bluebrother^hmm. At least no reds.
21:18:35bluebrother^but now the binaries are notably bigger than before.
21:19:00bluebrother^guess I have to check all linker options :(
21:21:43*bluebrother^ summons jhMikeS
21:22:12[Saint]Whoah....nice deltas :-S
21:22:17 Join kaputnik__ [0] (~kaputnik@port-92-206-79-54.dynamic.qsc.de)
21:23:18bluebrother^yes
21:23:30bluebrother^but at least it doesn't crash the mini2g directly on startup anymore
21:25:02[Saint]True - but at the cost of ~2000b deltas, is it worth it?
21:25:10bluebrother^yes :)
21:25:14 Quit kaputnik__ (Read error: No route to host)
21:25:15bluebrother^a crashing player is useless
21:26:00bluebrother^so for me two questions remain: (1) why has -Wl,−−gc-sections been added for all builds?, (2) what is the root cause for the problem we've been seeing?
21:26:30 Join amayer [0] (~amayer@72.25.43.33)
21:26:58bluebrother^I'm suspecting gcc to only drop objects that aren't used during linking. Since libfixedpoint.a includes only a single object it will include all functions since at least one is used.
21:27:06bluebrother^and thus causing the size delta
21:28:58 Join kaputnik [0] (~kaputnik@port-92-206-79-54.dynamic.qsc.de)
21:29:32bluebrother^we could of course put each function into a separate file, thus separate object. But that's kinda ugly ...
21:29:39bluebrother^and merely a workaround
21:30:17pamauryaren't we using -ffunction-section for that ?
21:30:51pamauryso that each function get its own section and the linker is hopefully smart enough to detect groups of sections which are unused ?
21:31:02bluebrother^I think we do
21:31:48bluebrother^otoh, shouldn't we be able to use −−gc-sections in general?
21:33:20bluebrother^hmm, there was some talk about that on the ml
21:33:21pamauryyes, I don't really understand why using −−gc-sections produces a faulty binary
21:35:29bluebrother^ok, so it seems we're at the point described here: http://www.rockbox.org/mail/archive/rockbox-dev-archive-2013-04/0014.shtml
21:36:05bluebrother^and following the answer http://www.rockbox.org/mail/archive/rockbox-dev-archive-2013-04/0015.shtml produces a broken binary, at least for mini2g
21:38:47pamaurymaybe that is a compiler/linker bug
21:42:25 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
21:42:43bluebrother^not sure
21:42:47bluebrother^"First of all, we need the -Wl,−−gc-sections option for the linker, but make sure this is not a safe flag to pass globally, so you should never put it in your LDFLAGS in make.conf."
21:42:52bluebrother^http://blog.flameeyes.eu/2008/01/today-how-to-identify-unused-exported-functions-and-variables
21:43:29pamaurythe only thing we can go bad with −−gc-section is when you have code in sections which are not "connected" to main, this happen with vector table for example or assembly code if the assembler doesn't track code
21:44:53*bluebrother^ rebuilds once more
21:46:24bluebrother^/usr/lib64/gcc/arm-elf-eabi/4.4.4/../../../../arm-elf-eabi/bin/ld: Removing unused section '.text.data_abort_handler' in file '/home/dom/projects/rockbox-gerrit/build-ipodmini2g/lib/libunwarminder.a(safe_read.o)'
21:46:42bluebrother^is this safe?
21:46:56pamauryno
21:47:10pamaurywell, not really ^^
21:47:41bluebrother^and safe_read.S is ... an assembler file
21:48:18bluebrother^hmm. Why are you using separate sections for the functions in that file?
21:48:28pamauryactually the problem here is different: data_abort_handler is called from crt0.S usually
21:49:16bluebrother^can it be called in normal operating mode?
21:49:26bluebrother^or is that only called in error situations?
21:49:50pamauryso I think the problem is that the linker doesn't see that crt0.S calls this functions, or that it thinks the crt0.S function which calls it is not used, which can happen if it called from .vectors and .vectors is not KEEP() in linker file. But anyway this is only called on data abort and backtrace
21:50:24pamaurylet me check ipodmini2g's crt0.S
21:52:24bluebrother^changing the .section directives in safe_read.S to make the linker not drop that anymore still produces a binary that doesn't work
21:52:55pamaurythat's expected
21:53:03bluebrother^too bad :)
21:53:12pamauryI suspect the problem is more about other handlers, like irq
21:53:39bluebrother^so you think the linker drops other stuff from crt0.S?
21:54:01pamauryif it drops data_abort_handler, I don't see why it wouldn't drop other things :-/
21:54:27*bluebrother^ wonders if it would be a good way to simply not use −−gc-functions and simply put each function in fixedpoint.c into a separate file
21:54:33pamauryexcept if this target doesn't use the backtracer of course
21:54:36bluebrother^kinda ugly :/
21:55:09pamauryI think i'll need to build the ipodmini2g myself to see
21:55:20bluebrother^http://pastebin.com/MyiD3TUL
21:55:37bluebrother^that's what the linker reports as getting removed
21:58:53pamaurynothing suspicous I would say
21:59:08pamaurydoes the crash address make any sense ?
22:00
22:10:15gevaertsI'm fairly sure I saw some deltas when the original patch went in, but I can't remember the details. I wouldn't be surprised if these deltas just reverse that
22:11:08gevaertshmmm
22:11:16gevaertsRight, no
22:11:32*gevaerts notices that the original is recent enough to still be in the table
22:11:43gevaertsWell, not entirely, anyway
22:12:01bluebrother^gevaerts: I had the same hope at first :)
22:12:31bluebrother^but right now I think it's safe to say that using −−gc-sections globally is unsafe.
22:13:37gevaertsIndeed
22:14:34gevaertsSo my guess is that fixedpoint.c has some stuff we don't always need
22:14:44bluebrother^yes.
22:14:49gevaertsIf that's the case, some ifdefs in there should do the trick
22:15:29 Quit tjb0607 (Quit: Segmentation fault)
22:15:42bluebrother^plugins link their own copy of fixedpoint, right?
22:15:56bluebrother^f.e. fp14_cos() is used in plugins only.
22:16:01gevaertsThat's what I understand, yes
22:16:08*gevaerts vanishes
22:16:30bluebrother^but since those now we link against a global libfixedpoint.a using ifdefs kinda doesn't work
22:16:37bluebrother^unless we build it twice
22:17:09*bluebrother^ ponders splitting fixedpoint.c up into multiple files again
22:18:21 Quit melmothX (Quit: #)
22:18:48 Quit bp0 (Quit: Leaving)
22:19:58bluebrother^hmm. fp_factor() and fp_decibels() could also be macros. Is there a reason against that?
22:28:44 Join jhMikeS [0] (~jethead71@50.4.247.132)
22:28:44 Quit jhMikeS (Changing host)
22:28:44 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
22:28:50funmani prefer static inline
22:29:22 Quit ranmachan (Read error: Operation timed out)
22:29:30 Join ranmachan [0] (ranma@kagami.uguu.de)
22:31:19 Quit simabeis (Ping timeout: 268 seconds)
22:31:27 Join simabeis [0] (~simabeis@lobmenschen.de)
22:31:40 Quit nomada (Ping timeout: 245 seconds)
22:33:31 Join nomada [0] (~nomada@irc.consciente.de)
22:35:28 Quit dv__ (Ping timeout: 264 seconds)
22:35:48bluebrother^fair enough
22:36:26bluebrother^hmm. fixedpoint.make doesn't handle more than one source file?
22:39:02 Join prof_wolfff [0] (~prof_wolf@62.83.50.196.dyn.user.ono.com)
22:40:31jhMikeSbluebrother^: when does it crash, btw. disk_unmount seems an odd place and nothing seems out of the ordinary in the .map file (no strange discards)
22:43:10*jhMikeS also wonders why he hasn't had issues, maybe should try h10 since it's the most similar thing
22:43:57bluebrother^jhMikeS: it's caused by −−gc-functions
22:44:31jhMikeSipod mini g2, correct?
22:44:37bluebrother^I've changed that for now and then figured that unnecessary code from fixedpoint.o is linked, increasing the binary size :/
22:44:40bluebrother^yes
22:44:47bluebrother^that's what it has been reported on
22:45:14jhMikeSthe .map says no
22:45:51bluebrother^huh?
22:46:11jhMikeSI might be out of sync here in the discussion
22:46:21bluebrother^seems :)
22:46:47bluebrother^I've removed setting −−gc-functions in fixedpoint.make a bit earlier tonight
22:46:54bluebrother^which makes mini2g at least work again
22:46:56jhMikeSah, yeah, without that, it'll include the entire lib. why must every target be adjusted?
22:47:23bluebrother^target adjusted?
22:47:59jhMikeSto not use −−gc-sections?
22:48:22bluebrother^I was reading the problematic diff and figured that we're now using −−gc-sections for the rockbox.elf link
22:49:10bluebrother^no, not every target. But if fixedpoint.make causes −−gc-functions to be set it will always be set if fixedpoint.a is linked
22:49:48bluebrother^but since −−gc-functions has effects on other linked libs (and the objects linked) I don't think it's a good idea to have a linked library enable −−gc-functions
22:50:53jhMikeS-ffunction-sections and -fdata-sections won't work then since the unused sections won't be discarded
22:52:56 Quit mrtux (Quit: WeeChat 0.4.0)
22:53:59jhMikeSlibs that aren't built with those flags should stay intact when linking if any part is used since the functions and data aren't distinguished by sections
22:58:55 Quit amayer (Ping timeout: 256 seconds)
22:58:56jhMikeS−−gc-sections will discard asm functions with sections like .text.functionname
23:00
23:03:19 Quit kevku (Ping timeout: 245 seconds)
23:04:14jhMikeSbluebrother^: was it accessing bad data or something else? perhaps and alignment problem?
23:06:33bluebrother^I haven't figured that :(
23:07:43 Join mrtux [0] (~mrtux@unaffiliated/mrtux)
23:08:07***Saving seen data "./dancer.seen"
23:10:51jhMikeSFS #12857 also reports a mini. hmm
23:10:51fs-bluebothttp://www.rockbox.org/tracker/task/12857 Data abort error on iPod Mini 2G (bugs, unconfirmed)
23:11:48 Quit mrtux (Client Quit)
23:13:04 Join amayer [0] (~amayer@h140.181.123.208.dynamic.ip.windstream.net)
23:15:04 Join mrtux [0] (~mrtux@unaffiliated/mrtux)
23:17:21 Join lebellium_ [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
23:18:12bluebrother^jhMikeS: I cannot reproduce the problem with 08199cd on H10. It's always reproducable on the mini2g though
23:18:30bluebrother^maybe related to the fact that the H10 is PP5020 while the mini2g is PP5022?
23:18:31 Quit lebellium (Ping timeout: 272 seconds)
23:18:31 Nick lebellium_ is now known as lebellium (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
23:22:09jhMikeSbluebrother^: my e200 works fine, also H10 does (but's my only P502x with an actual disk)
23:23:55jhMikeSI also don't see any discarded input sections that are not discarded in any other target
23:24:01[Saint]iPod 4G checks out.
23:24:12[Saint]Color/Photo/4G
23:24:57[Saint]I have one remaining with a spinning platter to test on. I left it so I could still test stuffs.
23:25:25[Saint]If I can do anything there - shout.
23:25:44*jhMikeS wants to check the .lds file
23:27:11*jhMikeS sees curious differences
23:30:30bluebrother^gtg for today. Can do tests tomorrow evening if necessary / useful. Just let me know ...
23:35:23 Join Cultist [0] (~Cultist@c-67-184-182-139.hsd1.il.comcast.net)
23:35:49CultistWill iPod Video themes work fine on the iPod Classic, since they have the same resolution?
23:36:34amayerCultist, yes
23:36:43Cultistsweet, thanks
23:38:47jhMikeS[Saint]: no mini g2? :)
23:39:03[Saint]No /functional/ mini g2, no ;)
23:39:14[Saint]though, I could probably build one....
23:41:03 Quit Wardo (Read error: Connection reset by peer)
23:42:43 Join lebellium_ [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
23:43:18 Quit n1s (Quit: Ex-Chat)
23:44:15 Quit lebellium (Ping timeout: 256 seconds)
23:44:24 Nick lebellium_ is now known as lebellium (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)

Previous day | Next day