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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2010-08-30

00:00:24gevaertsv3 is only the default battery capacity and debug menu current reading though, it doesn't change anything actually important
00:02:37gevaertssoap: updated
00:08:11 Quit leavittx (Ping timeout: 258 seconds)
00:08:30soapTYVM
00:09:11soapForum thread update?
00:09:34 Join leavittx [0] (~leavittx@89.221.199.187)
00:10:20gevaertsok, boss :)
00:12:39 Join h3po1 [0] (~h3po@p4FDE9D59.dip0.t-ipconnect.de)
00:15:14 Part domonoky
00:15:15 Quit h3po (Ping timeout: 276 seconds)
00:16:46 Quit hebz0rl (Quit: Ex-Chat)
00:21:08 Join bunnyboi [0] (~androgyne@cpe-72-224-19-157.nycap.res.rr.com)
00:21:36 Quit S_a_i_n_t (Ping timeout: 258 seconds)
00:22:49 Join S_a_i_n_t [0] (S_a_i_n_t@203.184.1.42)
00:23:02 Quit bunnyboi (Client Quit)
00:23:53 Join bunnyboi [0] (~androgyne@cpe-72-224-19-157.nycap.res.rr.com)
00:24:50 Quit h3po1 (Quit: Leaving.)
00:25:25 Quit bunnyboi (Client Quit)
00:35:16 Part froggyman
00:36:15 Quit Judas_PhD (Quit: This is a quitting message)
00:37:07 Join [Saint] [0] (S_a_i_n_t@203.184.0.140)
00:38:05 Quit kaiscene (Ping timeout: 258 seconds)
00:38:17 Quit S_a_i_n_t (Ping timeout: 272 seconds)
00:42:03 Join kaiscene [0] (kaiscene@ool-18e45d82.dyn.optonline.net)
00:42:57 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
00:43:05 Part JdGordon
00:50:05 Quit bertrik (Ping timeout: 265 seconds)
00:51:29 Join utanapischti [0] (~username@p4FF2CB04.dip.t-dialin.net)
00:52:01 Quit ender` (Quit: If I had only finished this sentence,)
00:54:16***Saving seen data "./dancer.seen"
00:55:23 Quit sasquatch (Ping timeout: 272 seconds)
00:56:15 Quit bluebrother (Ping timeout: 255 seconds)
00:57:50 Join bluebrother [0] (~dom@f053152218.adsl.alicedsl.de)
00:57:50 Quit bluebrother (Changing host)
00:57:50 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
01:00
01:01:32 Quit bieber (Ping timeout: 240 seconds)
01:01:50 Join bieber [0] (~quassel@162-78.97-97.tampabay.res.rr.com)
01:08:27 Join t0rc [0] (~t0rc@unaffiliated/t0rc/x-5233201)
01:23:07 Quit t0rc (Ping timeout: 245 seconds)
01:24:59 Join t0rc [0] (~t0rc@unaffiliated/t0rc/x-5233201)
01:27:40 Quit Zarggg (Quit: Zarggg)
01:29:02 Join Zarggg [0] (~zarggg@70.44.12.90.res-cmts.sm.ptd.net)
01:41:50 Join fdinel [0] (~Miranda@modemcable235.127-131-66.mc.videotron.ca)
01:48:52 Quit pamaury (Ping timeout: 240 seconds)
01:55:55 Join angelwolf71885 [0] (chatzilla@cpe-173-171-133-66.tampabay.res.rr.com)
01:59:26 Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
02:00
02:00:43 Join Ramsey[LC] [0] (~RamseyLC]@95.211.87.137)
02:03:07 Join JdGordon| [0] (~jonno@vl10.gw.ok-labs.com)
02:03:07 Quit JdGordon| (Changing host)
02:03:07 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
02:03:13 Part JdGordon|
02:06:30 Quit kaiscene ()
02:18:30 Quit [Saint] (Ping timeout: 272 seconds)
02:20:48 Join Strife1989 [0] (~Strife89@adsl-67-55-126.mcn.bellsouth.net)
02:21:17 Quit Dreamxtreme (Quit: I was raided by the FBI and all I got to keep was this lousy quit message!)
02:21:47 Join Dreamxtreme [0] (Dreamxtrem@92.30.98.82)
02:22:47 Join piotrekm [0] (~Adium@unaffiliated/piotrekm)
02:22:52 Join S_a_i_n_t [0] (S_a_i_n_t@203.184.2.128)
02:22:58 Part piotrekm
02:23:24 Quit Dreamxtreme (Client Quit)
02:23:32 Quit Strife89 (Ping timeout: 240 seconds)
02:24:09 Join Strife89 [0] (~Strife89@adsl-80-156-85.mcn.bellsouth.net)
02:25:07 Quit Strife1989 (Read error: Connection reset by peer)
02:34:52 Join Dreamxtreme [0] (~Dreamxtre@92.30.98.82)
02:35:03 Quit Dreamxtreme (Read error: Connection reset by peer)
02:47:03 Quit Zarggg (Quit: Zarggg)
02:47:45 Join [Saint] [0] (S_a_i_n_t@203.184.2.162)
02:49:34 Quit S_a_i_n_t (Ping timeout: 258 seconds)
02:54:18***Saving seen data "./dancer.seen"
02:58:00 Join webguest412 [0] (~cf664057@giant.haxx.se)
02:58:04 Join Dreamxtreme [0] (~Dreamxtre@92.30.98.82)
02:58:56webguest412ping
02:59:01Dreamxtremepng
02:59:03Dreamxtremepong
02:59:34 Quit ved (Ping timeout: 240 seconds)
03:00
03:01:24webguest412first time install of rockbox on ipod 5.5 80gb. Used the utility for auto install. Seemed to go well but on reboot can't find rockbox.ipod, but I checked and it is the <root>/.rockbox with a size of 711120 bytes. Any suggestions?
03:02:25webguest412install was from linux
03:12:23 Quit GeekShadow (Quit: The cake is a lie !)
03:29:34 Quit Strife89 (Quit: Good night, everyone. o/)
03:29:47 Quit _s1gma (Quit: irl ragequit)
03:43:22 Join Zarggg [0] (~zarggg@70.44.12.90.res-cmts.sm.ptd.net)
03:47:39 Quit angelwolf71885 (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716])
04:00
04:13:15 Join steve|m1 [0] (~steve@p4FD46304.dip.t-dialin.net)
04:13:15 Quit steve|m (Disconnected by services)
04:13:46 Nick steve|m1 is now known as steve|m (~steve@p4FD46304.dip.t-dialin.net)
04:25:23 Quit amiconn (Disconnected by services)
04:25:26 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:25:45 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:26:03 Quit TheSeven (Ping timeout: 265 seconds)
04:26:42 Quit pixelma (Disconnected by services)
04:26:45 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:26:47 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:30:02 Join kaiscene [0] (kaiscene@ool-18e45d82.dyn.optonline.net)
04:30:42 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:33:10 Join Barahir_ [0] (~jonathan@frnk-590fcd22.pool.mediaWays.net)
04:34:31 Quit Barahir (Read error: Operation timed out)
04:44:54 Quit MethoS- (Remote host closed the connection)
04:54:20***Saving seen data "./dancer.seen"
05:00
05:00:00soapwebguest412, you have the standard partition layout and a FAT32 ipod?
05:02:53 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
05:05:05 Quit anewuser (Ping timeout: 276 seconds)
05:12:29webguest412soap: don't know what standard partion layout is. Was a hfs ipod. Followed instructions on wiki for converting (dd'ing new part table to device, then formatting with mkdosfs on linux
05:13:11[Saint]Nano2G, current build, able to reproduce a crash on skipping multiple tracks very easily. Always gives the same address, but, nothing like it in the .map file :/
05:13:12soaphmm - I don't recall seeing this problem before.
05:13:24 Nick [Saint] is now known as S_a_i_n_t (S_a_i_n_t@203.184.2.162)
05:18:57 Quit panni_ (Quit: ( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de ))
05:20:26 Join dangdoo [0] (~ger@group23.vcn.bc.ca)
05:25:11webguest412soap: do you think that doing a manual unzip of the firmware (rockbox-ipodvideo64mb-3.6.zip) would fix the problem. I ask because some where in the install section of the manual it says that things installed manually will not be detected by the rbutility. Don't want to screw anything up.
05:25:57webguest412that is "...screws things up anymore than they already may be":)
05:34:50 Quit ps-auxw (Ping timeout: 264 seconds)
05:37:20 Quit Horscht (Quit: Verlassend)
05:45:48 Join ps-auxw [0] (~arneb@p4FF7F4F2.dip.t-dialin.net)
05:48:13 Quit webguest412 (Quit: CGI:IRC (Ping timeout))
05:51:05 Quit dangdoo (Ping timeout: 276 seconds)
05:52:46 Join webguest412 [0] (~cf664053@giant.haxx.se)
05:53:03 Join dangdoo [0] (~ger@group19.vcn.bc.ca)
05:59:47 Quit S_a_i_n_t (Ping timeout: 240 seconds)
06:00
06:01:23 Join S_a_i_n_t [0] (S_a_i_n_t@203.184.2.162)
06:05:04 Quit bieber (Read error: Operation timed out)
06:07:37 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
06:23:44 Quit storm` (Ping timeout: 276 seconds)
06:25:44 Join storm` [0] (~lol@ip98-179-182-180.ri.ri.cox.net)
06:28:41 Join likemindead [0] (~Xmccracke@ip72-198-31-155.ok.ok.cox.net)
06:29:51S_a_i_n_tHmmmm, with the Nano2G "skipping multiple tracks == crash" thing, different builds give a different address, but, still there's nothing even remotely close to it in the .map file.
06:29:56S_a_i_n_tHow do I debug this?
06:31:16likemindeadFrom what I see on the site, the Gigabeat S series is pretty solid, yes?
06:36:15 Join [Saint] [0] (S_a_i_n_t@203.184.4.91)
06:36:15 Quit S_a_i_n_t (Read error: Connection reset by peer)
06:36:30[Saint]if you want a DAP you can bash someone to death with, yes. the Gigabeats *are* very "solid" :)
06:38:00[Saint]If by "solid" you mean "stable" then, yes.
06:39:39likemindeadSweet. I found a 60GB S series for $50. :D
06:39:54likemindeadMy old iPod finally died after three years of Rockbox awesomeness.
06:52:20 Quit JdGordon (Ping timeout: 276 seconds)
06:54:24***Saving seen data "./dancer.seen"
07:00
07:05:36Jmaxlikemindead: i have a Toshiba Gigabeat S MESV60K and it works great with rockbox
07:07:24likemindeadHuh... Yeah, I couldn't figure out why it's an "unstable" port.
07:07:29likemindead(From the site.)
07:10:06 Join froggyman [0] (~me@unaffiliated/froggyman)
07:11:39 Join saratoga [0] (9803c22e@gateway/web/freenode/ip.152.3.194.46)
07:15:37 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
07:15:56JdGordon|anyone else getting "stkov power" with latest builds?!
07:16:01*JdGordon| never seen this before
07:19:42 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
07:20:49 Join MrSaturn [0] (~Quintin@c-24-128-227-217.hsd1.ma.comcast.net)
07:22:12 Part likemindead ("OBEY.")
07:22:17 Quit Chronon (Ping timeout: 272 seconds)
07:23:25MrSaturnHey, ive had my sansa e200 V1 booted with RB 3.3 for a while now, and i just picked it back up after a month or two, as i finally got a new SD card, but it keeps crashing... the light will still time out, and hold works, but sometimes ill just lose my ability to navigate, would anyone have any ideas why...?
07:23:43MrSaturni didnt feel like making a forum topic so i came here >_>
07:25:14 Part JdGordon|
07:25:44MrSaturnehh, ill just make a topic for this actually, but tomorrow, but if anyone wants to throw an idea at me, ill leave it open to come back to... i was thinking ill just put on the newest firmware
07:28:17saratogabefore you ask in the forums update to the latest build and see if that works
07:29:04[Saint]JdGordon: any idea why the themeeditor thinks
07:29:08[Saint]%?ic<%s%al%t(6)Composer:|%t(0)>;%?iA<%s%al%t(6)Alb Artist:|%t(0)>;%?ia<%s%al%t(6)Trk Artist:|%t(0)>
07:29:08[Saint]%?ic<%s%al%t(6)%ic|%t(0)>;%?iA<%s%al%t(6)%iA|%t(0)>;%?ia<%s%al%t(6)%ia|%t(0)>
07:29:11[Saint]is one line?
07:29:41[Saint]it displays as one line unless I add a newline between the two.
07:30:09[Saint]%?ic<%s%al%t(6)Composer:|%t(0)>;%?iA<%s%al%t(6)Alb Artist:|%t(0)>;%?ia<%s%al%t(6)Trk Artist:|%t(0)>
07:30:10[Saint]%?ic<%s%al%t(6)%ic|%t(0)>;%?iA<%s%al%t(6)%iA|%t(0)>;%?ia<%s%al%t(6)%ia|%t(0)>
07:30:15[Saint]argh...sorry.
07:30:22[Saint]didn't mean to paste that again.
07:42:56 Quit l403 (Read error: Connection reset by peer)
07:57:00 Join stoffel [0] (~quassel@p57B4DEDA.dip.t-dialin.net)
08:00
08:05:14 Quit saratoga (Quit: Page closed)
08:25:17 Join solexx_ [0] (~jrschulz@e176113250.adsl.alicedsl.de)
08:29:40 Quit t0rc (Ping timeout: 265 seconds)
08:31:52CIA-81New commit by 03amiconn (r27944): libdemac: ARMv7 assembler optimisation for the filters, tested on Nokia N900. Speedup is 2.1x for -c5000 compared to the ARMv6 asm. Note that actually ...
08:32:19 Join Guest87831 [0] (~bjst@rockbox/developer/Zagor)
08:33:04 Join LinusN [0] (~linus@giant.haxx.se)
08:33:04 Quit LinusN (Changing host)
08:33:04 Join LinusN [0] (~linus@rockbox/developer/LinusN)
08:33:39CIA-81r27944 build result: All green
08:40:10amiconn^Actually committed from N900 :)
08:41:20LinusNamiconn: respect! :-)
08:46:34 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
08:47:22 Nick Guest87831 is now known as Zagor (~bjst@rockbox/developer/Zagor)
08:49:41 Quit [Saint] (Read error: Connection reset by peer)
08:49:58 Join ender` [0] (krneki@foo.eternallybored.org)
08:50:41 Join l403 [0] (~l@gw8.vslesy.cz)
08:52:57 Join S_a_i_n_t [0] (S_a_i_n_t@203.184.1.132)
08:54:25***Saving seen data "./dancer.seen"
08:54:50 Quit dangdoo (Ping timeout: 258 seconds)
08:56:34 Quit webguest412 (Quit: CGI:IRC (Ping timeout))
09:00
09:03:26 Join leavittx_ [0] (~leavittx@89.221.199.187)
09:05:26 Join webguest412 [0] (~cf66404d@giant.haxx.se)
09:06:27 Join petur [0] (~petur@rockbox/developer/petur)
09:25:04S_a_i_n_tanyone else having problems "svn up" ing?
09:25:41S_a_i_n_tgetting "svn: Unknown Hostname 'rockbox.org' " :/
09:26:05S_a_i_n_tcsn't seem to ping it either....weird.
09:27:32 Join Rob2223 [0] (~Miranda@p4FDC9557.dip.t-dialin.net)
09:30:39gevaertswebguest412: did you use mkfs.vfat for this?
09:31:39 Quit Rob2222 (Ping timeout: 272 seconds)
09:34:39 Join einhirn [0] (~Miranda@p54850AA4.dip0.t-ipconnect.de)
09:43:31 Quit bmbl (Ping timeout: 258 seconds)
09:48:58 Quit mikroflops (Ping timeout: 265 seconds)
09:54:33 Join T44 [0] (~Topy44@f048005134.adsl.alicedsl.de)
09:58:36 Quit Topy44 (Ping timeout: 265 seconds)
10:00
10:12:43 Join petur2 [0] (~petur@rockbox/developer/petur)
10:13:54 Quit petur (Disconnected by services)
10:13:59 Nick petur2 is now known as petur (~petur@rockbox/developer/petur)
10:18:45 Join hebz0rl [0] (~hebz0rl@dslb-088-065-209-141.pools.arcor-ip.net)
10:30:39 Join Vampire [0] (~Vampire@unaffiliated/vampire)
10:40:14 Quit Judas_PhD (Quit: This is a quitting message)
10:41:26 Part Vampire
10:50:06 Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
10:54:16 Join pamaury [0] (~quassel@sal63-1-82-243-96-220.fbx.proxad.net)
10:54:17 Quit pamaury (Changing host)
10:54:17 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:54:29***Saving seen data "./dancer.seen"
11:00
11:06:36 Quit MrSaturn (Quit: Leaving.)
11:10:37 Part JdGordon
11:19:45 Quit swilde (Quit: ERC Version 5.3 (IRC client for Emacs))
11:23:43 Quit einhirn (Read error: Connection reset by peer)
11:28:03 Quit utanapischti (Quit: WeeChat 0.3.2)
11:28:27 Join sasquatch [0] (~username@p4FF2D9CB.dip.t-dialin.net)
11:43:32webguest412gevaerts: mkdosfs -c -F 32 -n RckbxiPod -S 2048 /dev/sde2
11:52:58 Quit petur (Read error: Connection reset by peer)
11:52:59 Join petur2 [0] (~petur@rockbox/developer/petur)
11:53:25 Nick petur2 is now known as petur (~petur@rockbox/developer/petur)
12:00
12:01:37 Quit webguest412 (Quit: CGI:IRC)
12:01:45 Join webguest412 [0] (~cf66404c@giant.haxx.se)
12:03:20 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
12:04:24 Join zenman [0] (~chatzilla@ppp118-208-135-34.lns20.bne1.internode.on.net)
12:06:51 Join mikroflops [0] (~yogurt@90-230-97-182-no112.tbcn.telia.com)
12:14:16 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
12:20:26 Join dfkt [0] (dfkt@unaffiliated/dfkt)
12:23:28 Quit Judas_PhD (Quit: This is a quitting message)
12:27:26zenmanhi, I'm new here. I have an iriver H340, and I'm running Ubuntu. I'm looking at the iriverboot page now for the firmware file - but how do I know which one I need?
12:34:36 Quit JdGordon (Ping timeout: 255 seconds)
12:47:28 Join _s1gma [0] (~d.d.derp@77.107.164.131)
12:47:46 Quit _s1gma (Max SendQ exceeded)
12:49:01 Join JdGordon [0] (~jonno@111.220.235.135)
12:49:08 Quit JdGordon (Changing host)
12:49:09 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
12:50:41 Join webguest25 [0] (~791552ed@giant.haxx.se)
12:51:56 Quit webguest25 (Client Quit)
12:54:30***Saving seen data "./dancer.seen"
12:55:04 Join Jerom [0] (~jerome@95.171.131.249)
12:57:44*JdGordon wonders why sdl+mini2440 touchscreen dont want to play nicely together
13:00
13:11:58 Join h3po [0] (~h3po@p4FDE8368.dip0.t-ipconnect.de)
13:12:44 Quit zenman (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100723084720])
13:13:45 Join ack` [0] (~ack@mingbai.org)
13:17:14 Join fip [0] (~inti@wirenat2.strw.leidenuniv.nl)
13:19:14gevaertswebguest412: right. mkdosfs is the same as mkvs.vfat. Try mformat as described on http://www.rockbox.org/wiki/IpodManualRestore. mkdisfs/mkfs.vfat has a (somewhat) known incompatibility with rockbox on 80GB ipods
13:27:28 Quit hebz0rl (Quit: Ex-Chat)
13:27:38 Join h3po1 [0] (~h3po@p4FDEA330.dip0.t-ipconnect.de)
13:31:28 Quit h3po (Ping timeout: 276 seconds)
13:35:26 Quit Jerom (Quit: Leaving.)
13:41:45 Join h3po [0] (~h3po@p4FDE82A1.dip0.t-ipconnect.de)
13:45:07 Quit h3po1 (Ping timeout: 276 seconds)
13:49:20 Quit l403 (Ping timeout: 240 seconds)
13:51:35 Quit JdGordon (Quit: Leaving.)
13:56:18 Join h3po1 [0] (~h3po@p4FDEA277.dip0.t-ipconnect.de)
13:58:07 Quit h3po (Ping timeout: 276 seconds)
13:59:32 Quit user890104 ()
14:00
14:01:12 Join MrSaturn [0] (~Quintin@c-24-128-227-217.hsd1.nh.comcast.net)
14:04:05TheSevenwebguest412: btw, volume names may be up to 11 chars, not 8 :)
14:11:49 Join h3po [0] (~h3po@p4FDEA633.dip0.t-ipconnect.de)
14:13:12 Quit MrSaturn (Read error: Connection reset by peer)
14:14:02 Join MrSaturn [0] (~Quintin@c-24-128-227-217.hsd1.ma.comcast.net)
14:14:22 Quit h3po1 (Ping timeout: 276 seconds)
14:25:44 Join h3po1 [0] (~h3po@p4FDEA08E.dip0.t-ipconnect.de)
14:26:50MrSaturnanyone know about how long it would take to uninstall my old build and put a new one on?
14:27:07soapnot long at all
14:27:22 Quit h3po (Ping timeout: 276 seconds)
14:27:52MrSaturnlike, less than 20 min? cause i have to leave soon
14:39:15MrSaturndoes anyone know if its bad to format your Mp3 player, cause it really seems to want me to >_>
14:41:18 Join h3po [0] (~h3po@p4FDE9F33.dip0.t-ipconnect.de)
14:41:39MrSaturnand if i should format, what file system?
14:42:03MrSaturnanybody? :(
14:42:32dfktwhat player are you talking about?
14:42:35soapWhy do you feel the need to format?
14:43:37 Quit h3po1 (Ping timeout: 276 seconds)
14:43:47MrSaturna sansa e200
14:44:13MrSaturnand it wont let me access my E;\ drive without doing so
14:46:13soap"it" being Windows?
14:46:23MrSaturnyes, windows 7
14:48:16MrSaturnso would it be safe do do so?
14:51:42soaphttp://www.rockbox.org/wiki/SansaE200Unbrick
14:52:12soapNot sure your player would be considered a "brick" - but the concepts discussed on that page are good for you to know.
14:52:47soapWhat happened prior to Windows 7 believing your sansa was unformatted?
14:53:10MrSaturnim not really sure
14:53:42MrSaturnall of my music was booted pre rockbox, and ive never had to access it since the first load
14:54:04MrSaturnplus its the first time ive used it in a few months
14:54:31***Saving seen data "./dancer.seen"
14:54:47MrSaturnand i wouldnt consider it bricked, as i can still play music, but ill definitely read that page, but my rides here, so i have to go anyways
14:55:04MrSaturnthanks for the help though soap!
14:55:28soapis your original firmware set to MTP or UMS mode? Files loaded via MTP won't be visible in UMS mode
14:55:58 Join h3po1 [0] (~h3po@p4FDEA424.dip0.t-ipconnect.de)
14:56:37 Quit h3po (Ping timeout: 276 seconds)
15:00
15:04:53 Join hebz0rl [0] (~hebz0rl@dslb-088-065-209-141.pools.arcor-ip.net)
15:06:07 Quit webguest412 (Quit: CGI:IRC (Ping timeout))
15:09:00 Join h3po [0] (~h3po@p4FDE9F2C.dip0.t-ipconnect.de)
15:10:09 Part LinusN
15:11:34 Quit h3po1 (Ping timeout: 276 seconds)
15:19:14 Quit krazykit (Quit: Connection reset by beer)
15:23:10 Join l403 [0] (~l@gw8.vslesy.cz)
15:25:16 Join h3po1 [0] (~h3po@p4FDE9D93.dip0.t-ipconnect.de)
15:27:10 Quit h3po (Ping timeout: 276 seconds)
15:27:37 Join panni_ [0] (hannes@ip-178-203-81-220.unitymediagroup.de)
15:33:33 Join jgarvey [0] (~jgarvey@cpe-065-190-066-089.nc.res.rr.com)
15:34:40 Quit MrSaturn (Quit: Leaving.)
15:39:37 Join krazykit [0] (~kkit@206.183.182.189)
15:40:33 Quit simonrvn (Quit: .)
15:40:45 Join h3po [0] (~h3po@p4FDE9C24.dip0.t-ipconnect.de)
15:42:46 Quit h3po1 (Ping timeout: 276 seconds)
15:55:46 Part fip
16:00
16:02:22 Join [Saint] [0] (S_a_i_n_t@203.184.0.253)
16:04:32 Quit S_a_i_n_t (Ping timeout: 265 seconds)
16:08:35preglowamiconn: hey, didn't you make your own optical to coax spdif converter?
16:08:56 Join funman [0] (~fun@rockbox/developer/funman)
16:21:55[Saint]Any iPod Video owners around to test a theme? http://www.datafilehost.com/download-7d0b8ce2.html
16:23:04[Saint].fms isn't really "up to par", but, general nit-picking is accepted.
16:23:22 Join ehntoo [0] (~ehntoo@lug.mtu.edu)
16:30:15 Quit leavittx (Ping timeout: 258 seconds)
16:30:38 Quit leavittx_ (Ping timeout: 258 seconds)
16:32:22 Part froggyman
16:33:10 Join leavittx_ [0] (~leavittx@89.221.199.187)
16:37:29 Quit AndyI ()
16:40:16 Quit leavittx_ (Ping timeout: 264 seconds)
16:43:38 Join leavittx_ [0] (~leavittx@89.221.199.187)
16:43:58 Join leavittx [0] (~leavittx@89.221.199.187)
16:44:47 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
16:50:45*rasher points kugel at http://www.reddit.com/r/Android/comments/d6zsi/rockbox_apks/
16:51:16 Join S_a_i_n_t [0] (S_a_i_n_t@203.184.0.89)
16:52:19 Quit [Saint] (Ping timeout: 276 seconds)
16:54:35***Saving seen data "./dancer.seen"
17:00
17:01:38 Join toffe82 [0] (~chatzilla@12.169.218.14)
17:04:53 Part Zagor
17:05:56preglowrasher: mwhat, we use sdl on android?
17:06:04rasherI thought we did
17:06:19rasherWe don't?
17:06:21preglowi don't think we do, but i could be wrong
17:07:26 Quit S_a_i_n_t (Ping timeout: 258 seconds)
17:07:32preglowi don't see why we'd want to do that, tho
17:07:44preglowthe native interfaces for everything is right here
17:10:05 Join S_a_i_n_t [0] (~st.lasciv@203.184.0.118)
17:10:43 Quit kaiscene (Ping timeout: 265 seconds)
17:11:11preglownah, it seems very sdl free
17:11:42*rasher edits
17:13:07preglowlet's hope i'm wrong then :>
17:16:02 Join kaiscene [0] (kaiscene@ool-18e45d82.dyn.optonline.net)
17:17:22 Quit kaiscene (Client Quit)
17:18:41 Quit MagusG (Read error: Connection reset by peer)
17:19:23 Join MagusG [0] (magusg@c-76-97-148-35.hsd1.ga.comcast.net)
17:28:03 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:28:24funmanno it doesn't use sdl
17:31:24 Join timccccc [0] (~lisa@121.159.100.28)
17:33:49 Quit petur (Quit: *plop*)
17:48:13 Quit S_a_i_n_t (Ping timeout: 272 seconds)
18:00
18:00:13funmanbluebrother: did you see http://www.rockbox.org/irc/log-20100828#20:22:03 ?
18:01:41 Join komputes [0] (~komputes@ubuntu/member/komputes)
18:02:19 Join S_a_i_n_t [0] (~st.lasciv@203.184.0.188)
18:03:21 Quit fxb__ (Ping timeout: 240 seconds)
18:03:45 Join esperegu [0] (~quassel@145.116.15.244)
18:07:35 Join evilnick [0] (0c140464@rockbox/staff/evilnick)
18:09:17 Quit S_a_i_n_t (Ping timeout: 241 seconds)
18:10:43amiconnpreglow: Yes I did, for both directions
18:18:00 Join user890104 [0] (Venci@Venci-Notebook-LAN.ipv6.6bez10.info)
18:20:12 Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
18:25:18 Join fxb__ [0] (~felixbrun@h1252615.stratoserver.net)
18:27:22 Quit panni_ (Quit: ( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de ))
18:46:46Not-iH8sn0wpie
18:50:49 Join Jaykay [0] (~chatzilla@p5DC5781A.dip.t-dialin.net)
18:51:03 Join CGL [0] (~CGL@190.207.141.245)
18:52:20preglowamiconn: got a schematic for it?
18:52:29preglowmy new sound card only has coax connectors
18:54:37***Saving seen data "./dancer.seen"
18:57:00 Join Lear [0] (chatzilla@rockbox/developer/lear)
18:59:04amiconnumm
18:59:13amiconnSomewhere, perhaps
18:59:48preglowmeh, can't seem to find female toslink connectors anywhere
19:00
19:00:37n1smurder an old dvd player
19:01:05amiconnYou need the TOTX173 for the coax->optical direction, and the TORX173 for the other
19:01:29 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:01:39amiconnThose thingies already have a standard toslink socket integrated
19:02:22preglowah, nice, i can get those
19:04:34amiconnFor optical->coax, you only need some resistors for level adjustment and impedance matching in addition to the TORX173
19:05:59amiconnFor the other direction I used a 74HCU04 for impulse forming. Oh, and you need regulated +5V
19:06:24preglowfigured i'd run it on batteries
19:06:27preglowhate cables :>
19:06:37*amiconn wonders whether he drew schematics, or just used some off-the-web schematics as-is
19:07:04preglowcobbling together a simple voltage regulator is simple anyway
19:10:06preglowhttp://www.beis.de/Elektronik/ADDA2496/SPDIF-Converter.html looks pretty decent
19:10:46bertrikpreglow, I think I found the person who provided the crossfeed algorithm, see the 2nd post in this thread http://www.hydrogenaudio.org/forums/index.php?showtopic=80428
19:11:19preglowjlohl is the guy
19:11:27amiconnYes, that looks usable, although it works a little different than mine
19:11:32preglowamiconn: how?
19:11:33 Quit timccccc (Ping timeout: 265 seconds)
19:12:20amiconnThis one lets you select optical or coax in, and outputs the selected signal on both coax and optical
19:12:51amiconnMine has no selector, just two fixed converters (coax->optical and optical->coax) on one board
19:13:27preglowwould be enough for me, so i'll probably strip out that part
19:13:52amiconnHmm, and he says the MAX3462 has problems at 96kHz
19:14:30preglowi can't get that part anyway
19:15:12amiconnHmm, it looks like the TO*X173 are only specced for up to 48kHz. My converter does work at 96kHz using those
19:15:50preglowi can get the alternate ic, tho
19:15:56 Join Kitr88 [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
19:16:12amiconnActually the switches are independent, but you still can leave them out
19:16:19bluebrotherfunman: thanks for the hint. Will create svn builds later so people can at least use Rockbox Utility again.
19:16:27preglowamiconn: the alternate receiver/transmitters are _really_ expensive, so no way i'm using those
19:17:00bluebrotherI'd like to check the open issues (and hopefully fix at least one) before making a new release, but if it turns out that I won't find the time for that I'm ok making a new release nevertheless.
19:17:12preglowamiconn: some spdif receivers are more tolerant of clock jitter than others, so it might just work with some gear
19:17:47bluebrotherI should be able to commit my lastest deployment script changes tonight. Still a bit ugly, but it seems to work fine at least.
19:18:48preglowbertrik: have you tried to calculate the typical speaker -> ear delay for some typical loudspeaker setups to see if 13 samples is far off?
19:19:02 Quit Kitar|st (Ping timeout: 276 seconds)
19:20:32 Quit Kitr88 (Ping timeout: 272 seconds)
19:21:17amiconnpreglow, bertrik: Just a thought - if we're ever going to support sample rate switching during playback, the crossfeed delay needs to become variable
19:21:19bertrikyes I have, 13 samples is equivalent to two speakers about 60 degrees apart
19:21:28gevaertsbluebrother: any timing on this new release? I'm hoping to get the ipod video merge done before next weekend, and I suspect a new Rockbox Utility would be helpful for that
19:21:51preglowamiconn: yeah, i know, but that can't be exactly supported without using very expensive fractional sample delays
19:21:58preglowamiconn: so there will always be some roundoff error in the delay time
19:22:00 Quit T44 (Read error: Connection reset by peer)
19:22:03bertrikso not really that far off, but a bit close to each other IMO
19:22:05bluebrothergevaerts: no timing yet. Since there wasn't any real changes to Rockbox Utility I didn't plan anything about that.
19:22:07 Join Topy44 [0] (~Topy44@f048005134.adsl.alicedsl.de)
19:22:36preglowbertrik: i think i've figured out a way to code variable delays in asm, btw
19:22:39bluebrotherbut if the merge is about to get finished the next week or so we should definitely try to not make a release 2 days before or so ;-)
19:22:48amiconnOf course there will be roundoff errors, but if 13 samples are about right to simulate speakers at 44.1kHz, this delay is way too much at 22.05kHz
19:23:14preglowamiconn: indeed, problem is the equivalent time delay difference between 6 and 7 samples is pretty big
19:23:43gevaertsbluebrother: ok. I think the merge patch is done, and I'm just waiting for some more testing, so that can probably be easily arranged
19:23:46preglowwe'd probably want something in between, but that would be hard to implement efficiently without destroying all high freqs
19:23:52 Join froggyman [0] (~me@unaffiliated/froggyman)
19:24:26 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:25:16 Join Kitar|st [0] (Kitarist@BSN-182-86-198.dial-up.dsl.siol.net)
19:25:33bertrikif you're playing back at 22 kHz then a crossfeed that is slightly off is probably not your biggest problem
19:25:45gevaertshm, should I post to General Discussions to point to my test build?
19:26:33 Join Kitr88 [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
19:28:27 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:28:47 Join [sko] [0] (~sko]@p57A992A4.dip0.t-ipconnect.de)
19:29:46 Quit Kitar|st (Ping timeout: 245 seconds)
19:30:45 Quit Kitr88 (Ping timeout: 240 seconds)
19:30:50 Join Kitar|st [0] (Kitarist@BSN-176-228-72.dial-up.dsl.siol.net)
19:31:37 Join Kitr88 [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
19:33:11 Join Buschel [0] (~chatzilla@p54A3AD89.dip.t-dialin.net)
19:34:13BuschelLear: you there?
19:34:18preglowi wonder if anyone ever made a fixed point implementation of an elliptic filter designer :>
19:34:45kugelrasher: what was that link supposed to tell me?
19:34:47 Join Jerom [0] (~jerome@95.171.131.249)
19:35:29rasherkugel: nothing in particular
19:35:38kugelah ok
19:35:43 Quit Kitar|st (Ping timeout: 265 seconds)
19:36:02kugelbut it shows that there's little point in releasing it in any way unless we make absolute point mode default
19:36:53 Quit funman (Quit: free(random());)
19:39:03 Quit Kitr88 (Ping timeout: 272 seconds)
19:42:36 Join ved2 [0] (ved@ddsbox.co.cc)
19:42:36rasherMakes sense on Android
19:42:59 Join Kitar|st [0] (Kitarist@89.142.57.91)
19:43:02*pixelma wants to draw kugel's attention to FS #11581
19:43:27 Nick ved2 is now known as ved (ved@ddsbox.co.cc)
19:46:06 Join Kitr88 [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
19:49:28 Quit Kitar|st (Ping timeout: 252 seconds)
19:51:05 Quit Kitr88 (Ping timeout: 272 seconds)
19:51:57CIA-81New commit by 03bluebrother (r27945): Support resolving of DLLs when running on Windows. ...
19:52:03CIA-81New commit by 03bluebrother (r27946): Update INSTALL file and move it in preparation of merging rbutil/ and utils/.
19:53:35CIA-81r27945 build result: All green
19:55:30CIA-81r27946 build result: All green
19:55:40 Join Kitar|st [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
19:56:17 Quit stoffel (Ping timeout: 276 seconds)
20:00
20:03:09 Quit powell14ski_ (Quit: powell14ski_)
20:04:13 Quit kugel (Remote host closed the connection)
20:08:16 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:14:14bertrikit's quite nice to read up a bit on the kind of audio filters used in rockbox
20:15:10 Join G4Oblivion [0] (~48281c8e@giant.haxx.se)
20:15:24preglowi don't think i use anything too fancy :>
20:17:12G4OblivionUpdate on nano 2g *PANIC* stkov nand: Using r27010 I do not have the problem. I will continue building different revisions with my free time untiI see the problem.
20:17:46 Quit G4Oblivion (Client Quit)
20:22:06 Join Kitr88 [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
20:22:36 Quit Kitar|st (Ping timeout: 272 seconds)
20:23:00LearYes.
20:33:42BuschelLear: you are able to play the "broken" forum-aac-file on your dap?
20:34:56LearBuschel: Yes, at least if it was fully buffered before starting playback of it.
20:35:22BuschelLear: I cannot. It stops with error ('Invalid number of channels') on iPod 5G
20:39:20LearBuschel: Is that an error code from faad?
20:39:21 Quit Kitr88 (Ping timeout: 245 seconds)
20:41:19BuschelLear: yes.
20:43:50kugelpixelma: I can reproduce but not yet tell why it happens :(
20:44:37 Join olejorgenb [0] (bronner@caracal.stud.ntnu.no)
20:45:12 Join Kitar|st [0] (Kitarist@BSN-182-111-88.dial-up.dsl.siol.net)
20:46:28kugelI tried e200v1 and fuzev1; in both cases the undefined instruction address is in the .idata section
20:48:16kugel(idata normally doesn't contain code, but initialized static variables)
20:49:08 Quit Jerom (Quit: Leaving.)
20:50:01 Join Jerom [0] (~heidi@95.171.131.249)
20:51:02LearBuschel: Pretty sure it is a buffering problem. If dircache is enabled, it plays just fine on an e200. With dircache off, it works if I play another track first (tested with an MP3).
20:51:44 Quit l403 (Ping timeout: 272 seconds)
20:54:41***Saving seen data "./dancer.seen"
20:55:21 Join Kitr88 [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
20:56:39preglowamiconn: you think electrolytic caps can be used as coupling caps on the coax line here? i don't remember if they work well with that kind of transients, and i happen to have a bunch of them i want to use :>
20:56:53 Join Buschel_ [0] (~chatzilla@p54A3AD89.dip.t-dialin.net)
20:58:13Buschel_Lear: I cannot reproduce, but I can playback the file when selecting it during buffering of another file...
20:58:48 Quit Kitar|st (Ping timeout: 240 seconds)
20:58:58Buschel_Lear: maybe the failure is connected to calling stream_seek() in libm4a/demux.c −− on the two problematic file I have, this is called
20:59:44Buschel_Lear: stream_seek is called, if the decoder needs to move back the file begin after reading at the file's end
20:59:50 Join dangdoo [0] (~ger@group26.vcn.bc.ca)
20:59:52LearYes, that happens when files aren't streamable (e.g., the "broken" file).
21:00
21:00:31 Quit Buschel (Ping timeout: 258 seconds)
21:00:36 Nick Buschel_ is now known as Buschel (~chatzilla@p54A3AD89.dip.t-dialin.net)
21:01:52preglowamiconn: and really, what's the whole point with the rs422 chip? is it just a voltage buffer or something?
21:03:05kugelhmm
21:04:00kugelchanging mpeg2_scan_norm/mpeg2_scan_alt to be in .ibss instead of idata (they're unitialized) gives a strange data abort instead of undefined instruciton
21:05:12amiconnpreglow: The rs422 chip is used as an amplifier/ buffer
21:05:19n1skugel: it sounds like it's executing some data or something
21:06:10amiconnThe TO*X173 input/ output at TTL level. Coax s/pdif is lower level and DC free
21:08:08amiconnFor optical -> coax you don't actually need an amplifier; my converter just uses a resistive divider with the proper output impedance and a acapacitor for blocking dc
21:09:33amiconnFor the other direction you do though - my converter uses the 74HCU04 for that (the buffered version wouldn't work well here)
21:10:28amiconnAnd I wouldn't use electrolytic capacitors for coupling s/pdif signals
21:10:59BuschelLear: hmm, any chance to further debug this issue?
21:12:27 Join MrSaturn [0] (~Quintin@c-24-128-227-217.hsd1.nh.comcast.net)
21:12:52preglowamiconn: but any other non-electrolytic cap type would do, right?
21:15:02 Join timccccc [0] (~lisa@121.159.100.28)
21:15:37kugeln1s: I'm assuming something goes wrong with initializing IRAM
21:16:03kugelmaybe due to linking order; I'm right now looking at the parameters passed to plugin_iram_init
21:20:34 Join dfkt_ [0] (~dfkt@unaffiliated/dfkt)
21:21:38 Join Kitar|st [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
21:21:52 Quit CGL (Ping timeout: 260 seconds)
21:23:53 Quit Kitr88 (Ping timeout: 276 seconds)
21:24:08 Quit dfkt (Ping timeout: 240 seconds)
21:25:48kugelbut I don't know if I can trust the .elf or .map files in this case :/
21:25:58 Quit Kitar|st (Ping timeout: 240 seconds)
21:26:43 Join Kitar|st [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
21:28:36 Join funman [0] (~fun@rockbox/developer/funman)
21:30:56LearBuschel: Enabling logf in buffering and playback (with perhaps a larger logf buffer) might give some ideas.
21:31:15 Quit Kitar|st (Ping timeout: 265 seconds)
21:32:42 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
21:33:05funmanbluebrother: are you aware of this kind of 'exploits' ? http://www.exploit-db.com/exploits/14750/ (i see you just made a commit about DLLs)
21:40:26 Nick fxb__ is now known as fxb (~felixbrun@h1252615.stratoserver.net)
21:42:26 Nick fxb is now known as fxb__ (~felixbrun@h1252615.stratoserver.net)
21:43:45 Join Buschel_ [0] (~chatzilla@p54A3AD89.dip.t-dialin.net)
21:44:52funmankugel: IIUC, the iram copy is memset() to zero because it's also the bss segment
21:45:15Buschel_Lear: will try to check the buffering code...
21:45:38kugelfunman: uhm, yea, possibly, I don't know how that helps me though
21:46:06 Join Kitar|st [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
21:46:21 Quit Buschel (Ping timeout: 240 seconds)
21:46:27 Nick Buschel_ is now known as Buschel (~chatzilla@p54A3AD89.dip.t-dialin.net)
21:46:33funmani can't find where bss is zeroed?
21:46:47funmanah, plugin__start
21:47:56 Quit n1s (Ping timeout: 276 seconds)
21:48:23LearBuschel_: The threading makes it tricky to debug though.
21:48:30funmandidn't you want to get rid of PLUGIN_IRAM_INIT ?
21:48:36kugelhm
21:48:53kugelgood point, I'll see if I don't overwrite iram copy
21:49:40kugelfunman: maybe doing that would fix it (in a better way)
21:49:51funmani'm not sure why codecs don't break though
21:50:10 Quit dangdoo (Quit: using sirc version 2.211+ssfe)
21:52:21 Quit Kitar|st (Ping timeout: 240 seconds)
21:53:27kugelfunman: I copy&pasted the bss zero'ing from the codecs actually
21:54:06kugel0x00000000307a7994 iramcopy = .
21:54:07kugel0x00000000307a7994 plugin_bss_start = .
21:54:37 Join Lear_ [0] (chatzilla@rockbox/developer/lear)
21:55:07funmanperhaps there's something special for codecs
21:56:48MrSaturnI see that you guys are doing something, but anyone want to help me troubleshoot? My Sansa e200 will not let me access it when its in MSC mode, but i can get in fine when its MTP mode...now i want to update rockbox, but to do so i have to be in MSC
21:57:45MrSaturnIt says the drive is corrupt and to reformat when im in MSC, trying to access it
21:57:53 Join Kitar|st [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
21:58:10 Quit Lear (Ping timeout: 252 seconds)
21:58:17 Nick Lear_ is now known as Lear (chatzilla@rockbox/developer/lear)
21:58:19kugelfunman: ah, the old code did memset(pluginbuf+readsize, ...) while memset(plugin_bss_start, ...) obviously doesn't
21:58:30kugelI thought it was equivalent but I didn't think of iramcopy
21:58:45funmankugel: where was the old code?
21:59:01kugelplugin.c (see diff of r27902)
21:59:27funmani don't get the difference
21:59:52kugelreadsize (return of read()) includes iramcopy because it's in the binary
21:59:59funmanoh
22:00
22:00:13funmanso bss is not 100% cleared until PLUGIN_IRAM_INIT() is run?
22:00:19kugelyes
22:00:31kugeleven worse, it contains code
22:00:34funmani'm trying to implement your solution adn get rid of PLUGIN_IRAM_INIT
22:00:54kugelso, doing something like plugin_running = true; from within the plugin before PLUGIN_IRAM_INIT could easily break
22:01:42kugelfunman: go ahead if you like, I'd do it too
22:01:54MrSaturnill try the forums >_>
22:02:13kugelI think the biggest problem is that iram* is undefined for plugins not using iram (but maybe it isn't)
22:02:29 Quit Kitar|st (Ping timeout: 264 seconds)
22:02:31funmanit's not but we can change it
22:03:05funmanhttp://pastie.org/1127376
22:03:21funmaniramend-iramstart and iend-iedata will be 0
22:07:38kugelgreat
22:08:16 Join clone4crw [0] (~calvin@96-42-90-88.dhcp.roch.mn.charter.com)
22:08:27funman /* Disable all talking before initializing IRAM */
22:08:31funman(from rockdoom.c)
22:08:51kugelanother question, do you know why cpucache_invalidate() is #if NUM_CORES > 1?
22:09:09kugelsurely single core builds also need cache coherency if code is copied around?
22:09:16funmanthat's right
22:09:45 Quit toffe82 (Read error: Connection reset by peer)
22:11:25amiconnkugel: DonMt forget that pluginiram usage is optional - and plugins usable during playback *must not* touch iram
22:11:44kugeli won't forget that
22:12:20funmancpucache_flush() is only exported in plugin API for dual core, but iiuc it should be used for every self modifying code
22:12:49amiconnSingle core not using disk dma and having unified cache doesn't need to care
22:13:08BuschelLear: with this patch the files play on my 5G -> http://fs10u.sendspace.com/processupload.html
22:13:13kugelbut most caches aren't unified (the standard arm one for example=
22:13:25funman+ there is no distinction in rockbox
22:13:26BuschelLear: but each seek leads to rebuffering in this case.
22:14:21kugelfunman: http://pastie.org/1127407 works for me
22:14:44LearBuschel: "An error occurred while processing the uploaded file."
22:14:47funmani think i have almost the same thing
22:15:26BuschelLear: http://pastie.org/1127411
22:15:29LearBuschel: A rebuffer is expected (unless the file was already buffered).
22:15:38funmanwell not exactly
22:15:44funmani was removing PLUGIN_IRAM_INIT also
22:16:01kugelso I do
22:16:06amiconnfunman: Do we have self modifying code in plugins?
22:16:15LearBuschel: Ah, I see what you mean now.
22:16:37funmanamiconn: all plugins using iram.
22:16:50amiconniram is uncached
22:16:55funmanno
22:17:32amiconnwell, at least it is on all targets that existed when these mechanisms were implemented
22:17:59amiconn(coldfire, pp,)
22:18:05 Join Kitar|st [0] (~Kitarist@BSN-210-255-49.dial-up.dsl.siol.net)
22:18:34kugelfunman: bubbles compiles and runs (without stopping playback) without your change to plugin.lds
22:18:37funmankugel: what's the diff between audio_stop() & audio_hard_stop() ?
22:18:57kugeldunno, I was just too lazy to add audio_hard_stop() to the api :)
22:19:05amiconnIt makes no sense to cache iram when it is as faster (or even faster due to a bug on PP5002) as the cache
22:19:19kugelbut I'm fairly sure that audio_stop doesn't return before audio is really stopped
22:19:20funmankugel: ah right, the .lds is common
22:20:15funman(to all plugins)
22:20:52kugelso your change isn't needed?
22:21:14funmanno
22:22:00kugelok, so I guess I run sed and then commit
22:22:12funmanhttp://pastie.org/1127435
22:22:24funmani'm not sure if const is needed
22:23:02 Quit Kitar|st (Ping timeout: 276 seconds)
22:23:33funmanit doesn't help removing the code from the .elf at least
22:23:43kugellooks good. the occurences of PLUGIN_IRAM_* still need to go
22:24:20kugelfunman: plugin_crt0.c is also common to all plugins (only compiled once)
22:24:28kugelit'd need lto to remove that code
22:24:29funmanoh
22:25:21funman#define HAVE_CPUCACHE_INVALIDATE
22:25:21funman#define HAVE_CPUCACHE_FLUSH
22:26:18BuschelLear: can it be connected to STORAGE_WANTS_ALIGN ?
22:26:38 Join bunnyboi [0] (~androgyne@cpe-72-224-19-157.nycap.res.rr.com)
22:26:39LearDon't know, but I doubt it.
22:26:42funmanalso we need a memory barrier between iram copy & bss clearing?
22:27:02kugelamiconn: for newer ARMs the iram is rather slow so it makes sense to cache it, and the cache is split for code and data so invalidating is needed
22:27:07 Quit bunnyboi (Read error: Connection reset by peer)
22:27:28kugelfunman: not if plugin_crt0 does handle it properly
22:27:37 Join {phoenix} [0] (~dirk@p57AA6014.dip.t-dialin.net)
22:27:52funmaniiuc iramcopy & bss clearing could be reordered, because gcc doesn't know they use the same location?
22:28:22 Quit ender` (Quit: Kiss a pair of boobs and the movie's rated R. Chop them off and it's PG-13.-- Jack Nicholson (?))
22:29:09kugelhm that seems rather unlikely
22:29:19kugelhow would a memory barrier look like?
22:29:39preglowBuschel: you wouldn't have any idea about how to do cheap and good(ish) resampling from 48khz to 44.1khz? :>
22:29:41kugelcould a goto prevent reordering?
22:29:41funmanasm("" ::: "memory"); i think
22:30:30Buschelpreglow: not really
22:30:33funmani don't see why
22:31:05kugelthe compiler must not put code that's before the label after the label
22:31:16kugelbut if that little asm makes things safe go for it
22:31:24pregloweverone and their dog seems to love using polyphase fir filters
22:31:30preglowwhich aren't exactly cheap
22:31:43preglowand love to hog memory
22:32:14 Join h3po1 [0] (~h3po@p4FDE983B.dip0.t-ipconnect.de)
22:32:32gevaertskugel: why mustn't a compiler do that?
22:32:58 Nick dfkt_ is now known as dfkt (~dfkt@unaffiliated/dfkt)
22:33:00funmankugel: http://pastie.org/1127465
22:33:07kugelbecause it would jump before the label, not after?
22:33:09 Join simonrvn [0] (simon@211.148-ppp.3menatwork.com)
22:33:29gevaertsyes, but it knows where the jump comes from, so it could add the same code there
22:33:35 Quit h3po (Ping timeout: 276 seconds)
22:34:00kugelright
22:34:10kugelfunman: is _flush correct?
22:34:43funmanhum no..
22:34:54funmaniiuc it should be invalidate so the memory is fetched again
22:35:14kugelI think so too
22:35:29amiconnfunman: Make that 'asm volatile'
22:36:20funmannot sure why that was needed on PP at all?
22:36:26solexx_Playback stopped working a few days ago on my iriver H120. SVN from 08/08 still works (got it as RAM image), current version doesn't.
22:37:31funmankugel: cpucache_invalidate might be needed not only if we use iram also
22:38:13kugelwhy?
22:38:23funmanloading plugins is self modifying code
22:38:31kugelah yes
22:38:43kugelbut lc_load() does _invalidate() already
22:39:13 Quit [sko] (Quit: Leaving.)
22:39:30funmanhm and cpucache_invalidate() is called unconditionally
22:39:46funmanis there a target which doesn't define it?
22:40:11kugelI think it's an empty static inline for all others
22:40:52funmanright, in system.h
22:41:20 Quit Jaykay (Ping timeout: 265 seconds)
22:41:55funmanshould i commit?
22:41:56bluebrotherfunman: I've read about this type of exploits recently.
22:44:01 Quit liar (Ping timeout: 258 seconds)
22:44:06kugelfunman: sure
22:44:09LearBuschel: Hm, is it the initial "avail" calculation in prep_bufdata that is the problem? If widx is smaller than ridx, it is seen as a buffer wrap. Wouldn't that be wrong if the data hasn't been buffered yet?
22:46:14solexx_Ah, forget it. Filesystem error.
22:46:29 Quit clone4crw (Quit: leaving)
22:47:11BuschelLear: I am not knowing much about the buffering code ... :/
22:47:31CIA-81New commit by 03kugel (r27947): Change lc_open_from_mem() return type to void*
22:47:56CIA-81New commit by 03funman (r27948): Plugins: modify IRAM copying code ...
22:49:16kugelfunman: good work, thanks :)
22:49:38CIA-81r27947 build result: All green
22:49:41funmankugel: is it a self-congratulation, because you made the same code? ;)
22:49:49kugelshh! :P
22:50:26LearBuschel: I don't know it well, but I've dug into it a bit before... But it seems to me prep_bufdata should go into the "wait for data" loop in this case, but doesn't, and reads garbage.
22:51:20LearBuschel: Maybe your "fix" should be applied, until a proper fix can be devised - preferably by someone who knows the code well. :)
22:51:45CIA-81r27948 build result: All green
22:52:02 Join clone4crw [0] (~calvin@96-42-90-88.dhcp.roch.mn.charter.com)
22:52:30bertrikI remember NicoP working on / fixing buffering.c, but I don't see him around much anymore
22:54:29funmanlogbot doesn't remember him
22:54:45AlexPHe did metadata on buffer
22:54:46***Saving seen data "./dancer.seen"
22:54:55AlexPBut hasn't been around for a year or two now
22:56:21BuschelLear: But with my "fix" the playback will have dropouts due to permanent rebuffering. so, no solution either
22:56:55pixelmayes, and Nico_P went away with LinusN's Ipod Video
22:56:57LearPermanent rebuffer? Could you explain a bit more?
22:58:40CIA-81New commit by 03kugel (r27949): Update comments in the helloword plugin.
22:58:53BuschelLear: If I comment the code section like in the patch I have linked, the HDD is always spinning and the playback stucks each 1-2 seconds
22:59:27LearFor any file? Not just the problem file?
22:59:30kugelanyone fancy at looking into buffering? buffer_handle() is a bit broken (see the FIXME in it)
23:00
23:00:05BuschelLear: let me check (only tried for one of the problem files)
23:00:41CIA-81r27949 build result: All green
23:00:43AlexPbertrik: last commit 25/03/2009 :)
23:00:44LearHm, there'd be lots of forward seeks in the demuxer, true...
23:01:32Lear(Or maybe not, when I think a bit more about it.)
23:02:14funmankugel: looking into buffering is a punishment IMO :
23:02:16funman:)*
23:02:33kugelsome people like punishments :)
23:02:52pixelmakugel, funman: thanks both for looking into the crash problem
23:03:41Learkugel: Hey!
23:04:06funmanpixelma: you're welcome
23:04:14BuschelLear: interesting it is not happening with our "broken_m4a.m4a" but with another problem file ("Kaprsberger.m4a" from FS #11539)
23:04:35Buscheltoo many typos. need a nother beer ;)
23:04:56LearHm, what about this: in rebuffer_handle, line 782 should probably use ringbuf_add, and then wait for the buffer message to return. Or does queue_send do that already?
23:06:17kugelqueue_send() blocks
23:08:17preglowamiconn: do you happen to have a h1x0 hanging around close to your spdif interface?
23:08:34amiconnfunman: Hmm, now plugins have no chance to ask before stopping playback when they want to use iram. No plugin actually used this possibility though
23:10:09amiconnIt would have been potentially dangerous anyway, as it would have meant to run with uninitialized bss, and even worse, potentially overwriting iram content before it has been copied
23:12:12 Part MrSaturn
23:12:21kugelno loss of nothing used it IMO
23:12:30kugelmost iram plugins play their own sound anyway
23:12:36amiconnpreglow: Define 'your' s/pdif interface. My H180 is may main player when at home, but nowadays I don't need the converter because my hifi set does have both optical and coax s/pdif
23:13:21*amiconn mainly built it for testing purposes, as it converts nicely between H1x0 and Recorder
23:13:37funmanaudio_stop() doesn't stop playback?N
23:14:35funmanmpegplayer works fine for me
23:15:38 Join h3po [0] (~h3po@p4FDE9634.dip0.t-ipconnect.de)
23:15:50 Quit h3po1 (Ping timeout: 276 seconds)
23:16:02LearBuschel: Does the ipod video use DMA for HD? If so, the thread calling read() would go to sleep more or less immediately?
23:16:23BuschelLear: yes, DMA is used
23:17:24LearThat could explain why it is worse on that target. rebuffer_handle would return to early.
23:17:27 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
23:17:39 Quit jae (Remote host closed the connection)
23:17:43 Join jae [0] (~jae@jaerhard.com)
23:17:54amiconnHow so?
23:18:14amiconnThe read() call is still synchronous, just that other threads may run meanwhile
23:18:49LearYes, which means the codec thread gets to run (queue_send returns), trying to read data that hasn't been read yet.
23:20:03amiconnAfaiu queue_send() is synchronous as well
23:20:14kugelqueue_send doesn't return before the other thread has acked
23:20:31LearYes, but queue_reply() is called before read() in the buffer thread.
23:21:00 Quit dfkt (Quit: -= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.)
23:21:10kugelwhere?
23:21:31LearLine 1464 in buffering.c
23:22:38LearHm, but the "seek fix" works on the e200 too...
23:23:49BuschelLear: Good! So, you can reproduce it :)
23:24:20LearSure, but I can "fix" it by turning on the dircache.
23:25:36 Part domonoky
23:26:37preglowamiconn: i was basically just thinking of finding out how retailos does resampling to get some hints on how i should look into implementing it
23:26:43preglowamiconn: and i need spdif recording for that
23:26:45 Join S_a_i_n_t [0] (S_a_i_n_t@203.184.2.204)
23:27:21amiconnSo what would be required to play/ record to/ from where?
23:27:38preglowh1x0 playing 48khz wavs -> some spdif recording device
23:28:03amiconnThe only other s/pdif recording device I have apart from the H1x0 is my old trusty Recorder
23:28:17preglowis that able to record wavs, tho?
23:28:30amiconnYes, using the wavrecord plugin
23:28:38preglowshould do the trick, then
23:29:00preglowif i remember correctly, iriver resamples everything to 44.1
23:29:18amiconn(not as comfortable as in-core recording, but for s/pdif you don't have to care about recording level)
23:29:42 Quit jae (Ping timeout: 265 seconds)
23:29:54amiconnYes, it has to resample at least 48/32/24/16/12/8
23:30:13amiconnThe hardware only allows 44.1/22.05/11.025
23:30:36 Join jae [0] (~jae@jaerhard.com)
23:30:36amiconnOh, and the corresponding high rates of course
23:30:45 Quit hebz0rl (Quit: Ex-Chat)
23:31:02amiconn(96/64 needs resampling, 88.2 could be played natively)
23:32:36 Join h3po1 [0] (~h3po@p4FDE9519.dip0.t-ipconnect.de)
23:33:02 Quit esperegu (Remote host closed the connection)
23:33:25preglowi'll toss up some wav files you can record at your leisure
23:33:33preglowa simple sweep and some impulses
23:33:41preglowshould give me an idea of what they do
23:34:03 Quit h3po (Ping timeout: 276 seconds)
23:34:18*Buschel slaps his forehead
23:35:02BuschelLear: my hotfix is working. the dropouts were not caused by buffering issues
23:35:30BuschelLear: the other problematic file is an aac-he −− which is not playable realtime on my 5G :/
23:35:42Lear:)
23:36:44 Quit jae (Remote host closed the connection)
23:37:29 Quit timccccc (Read error: Operation timed out)
23:39:16BuschelShall I submit the dirty buffering patch? -> http://pastie.org/1127646
23:39:38preglowamiconn: http://www.pvv.org/~thomj/sweeptest.wav and impulsetest.wav
23:40:31LearBuschel: Don't know. I'm about to try a what I hope is a better fix...
23:41:10BuschelLear: btw, maybe this also fixes FS #11476 (some buffering issues with mpc, when using seek)
23:41:30BuschelLear: just let me know when there is something to test
23:41:49LearNope, my idea didn't work...
23:42:00 Join jae [0] (~jae@jaerhard.com)
23:45:14 Quit kramer3d (Quit: This computer has gone to sleep)
23:45:56 Quit user890104 ()
23:46:10kugelBuschel: what does the patch do?
23:46:21 Join user890104 [0] (Venci@Venci-Notebook-LAN.ipv6.6bez10.info)
23:46:26BuschelIt fixes an issue that is described in http://forums.rockbox.org/index.php?topic=25602.15
23:47:17BuschelIt may also be connected to FS #11476 −− I cannot verify this though...
23:48:39kugelso it assumes one chunk is buffered but that may not always be the case?
23:50:12*S_a_i_n_t wonders if anyone was kind enough to test the theme for iPod Video which he posted a few hours ago...
23:50:32BuschelAs far as I understood from the buffering code, the commented section avoids to rebuffer the full file, if only a small part (the next CHUNK_SIZE bytes) needs to be read.
23:51:03BuschelAnd this special behaviour (performance optimization?) is not working flawlessly
23:51:29kugelin what case does it cause problems?
23:51:55kugelI assume it's not necessarily useful for audio but maybe for other data on the buffer (metadata, albumart?)
23:52:23LearBuschel: And that's a special case too; h->offset is the offset from which buffering starts, so it is only done for seeks in the first chunk, before the first chunk has been read.
23:52:57BuschelJust give the file that is linked in http://forums.rockbox.org/index.php?topic=25602.15 a try. It is called "broken_m4a.m4a"
23:53:10kugelmaybe the queue_send() simply shouldn't be happen before read() has returned?
23:54:16BuschelLear: So, this special special case might not be needed?
23:54:19LearThere really should be a buffering going there, so maybe it'd be enough to wait for it to happen.
23:54:52 Quit bertrik (Quit: :tiuQ)
23:55:08BuschelI have added a flyspray entry for this -> FS #11586
23:55:09LearBuschel: Well, it would improve things during codec start, I imagine. But maybe not that much.
23:57:11BuschelLear: If so, I would prefer to comment this special behaviour and wait for a detailed analysis.
23:58:33 Quit krazykit (Quit: Connection reset by beer)
23:58:41kugelI wonder why it does queue_send() in that particular part. isn't the optimization about that it's already buffered?

Previous day | Next day