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 2009-10-01

00:00:44HillshumRoger__: Tell it to build the library, plug your device in, right click on music and tell it to sync
00:00:57robin0800Roger__: no I hate media monkey
00:01:07Zagorsure. I just need to check out all the files first.
00:02:23HillshumRoger__: This is getting off topic too. >> #rockbox-community ?
00:02:36 Join Stephen_ [0] (n=S@86.45.119.71)
00:03:05Roger__sorry, thanks
00:04:56tomersSometimes in the past someone told me about a cool utility to compare two patches. Anyone aware of any?
00:05:11Hillshumdiff?
00:06:01Torneinterdiff
00:06:03tomersNo. I worked on a vanilla code, apply a patch, and then added my own stuff. I want to see the diff between the current modified code to the original patch
00:07:02tomersTorne: Thanks. Seems to be it
00:08:56 Quit robin0800 (Remote closed the connection)
00:09:50 Quit gregzx ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
00:12:04Tornehm, i need a disassembly of ipod diag mode :)
00:12:05kugelhttp://www.realworldtech.com/page.cfm?ArticleID=RWT093009110932&p=1 - very interesting
00:12:09kugeloops
00:12:26*Torne thinks he's found the a2ds with the charger detect
00:12:50 Quit barrywardell ()
00:22:08Torneinteresting
00:22:28Tornethe pcf50605/6 datasheet says only the 50606 has ADC
00:22:57Torneyet the ipodvideo port does ADC while claiming to be a 50605
00:24:30 Quit bertrik ("De groeten")
00:27:37saratogaanyone willing to host test files for test_codec?
00:27:43saratogai doubt i have enough space for a comprehensive set
00:30:04Hillshumsaratoga: Zagor?
00:30:35Zagorhow much is it?
00:31:51saratogaZagor: 242MB
00:32:02Zagorno problem
00:32:10Zagorcan you make them?
00:32:13saratogaok let me upload what i can and i'll send you the link
00:34:17 Quit advcomp2019 (Read error: 113 (No route to host))
00:34:29 Quit LambdaCalculus37 ("Fwump")
00:35:31 Quit DerPapst ("Leaving.")
00:35:47 Quit domonoky (Read error: 104 (Connection reset by peer))
00:36:30 Quit mt_ ("Java user signed off")
00:37:12stripwaxTorne - does that adc also do accessory detect?
00:37:23***Saving seen data "./dancer.seen"
00:37:46TorneI don't think it's the same one, no
00:37:55TorneThere are a lot more adc values shown in diagmode than we read.
00:38:15Tornesuspect the hardware may be using all 5 adc channels?
00:38:27Tornewe only read 3 (well, 2 in svn, the third is from the charge current patch)
00:38:38Tornediagmode appears to have an adc hooked to D- and D+
00:38:47Torneand calculates a charging yes/no flag from those.
00:39:02Tornewhich is consistent with reports on the internet
00:39:16Tornei'm gonna throw diagmode into ida and take a look
00:39:32Tornesee if it's using the pcf for them all, and which channel/modes it uses
00:40:04 Quit Stephen_ ("Leaving")
00:40:22 Quit mt (Read error: 104 (Connection reset by peer))
00:40:26stripwaxTorne - what's the charge current patch?
00:40:42Tornefs#9728
00:40:51Torneshows current flowing into/outof battery on ipodvideo, maybe others
00:40:56Torneappears to eb accurate :)
00:41:01Torne(uses scaling factors from diagmode)
00:41:36stripwaxso more accurate charging / discharging status?
00:41:43stripwaxor just for info
00:41:59saratogaZagor: http://duke.edu/~mgg6/rockbox/files.zip.001
00:42:05stripwax(I see that patch is just for debug_menu)
00:42:10Torneit's just for info
00:42:19Tornethe patch just displays it in battery status
00:42:32saratogaZagor: this too: http://duke.edu/~mgg6/rockbox/files.zip.002
00:43:30stripwaxsaratoga - what codec(s) are these for? I'm thinking we should have some test-case files hosted somewhere too for semi-regular regression testing (e.g. those vorbis files that caused crashes on some gigabeats a month or two back)
00:43:41stripwax^should have^should also have (somewhere)
00:43:49saratogastripwax: most major ones
00:44:26stripwaxis it basically the output of that standard test file thru all the codec bundles, or a bunch of other stuff also?
00:44:36saratogathat and APE as well
00:45:10saratogaZagor: bah forget it those files come out corrupted anyway
00:45:19Zagoroops :)
00:45:43saratogalet me try sending via SSH to my school
00:46:50 Quit bmbl ("Bye!")
00:49:55saratogahttp://duke.edu/~mgg6/rockbox/files1/
00:49:57saratogafirst batch
00:50:11saratogawill send the others after dinner, let me know when you have them
00:52:12 Quit ender (" cd /pub && get beer")
00:58:18 Join Thundercloud [0] (i=thunderc@persistence.flat.devzero.co.uk)
00:59:32Zagorsaratoga: I have them. but I'm off to bed now. I'll grab the rest tomorrow morning.
01:00
01:00:00 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
01:00:24 Quit Zagor ("Clint excited")
01:02:34 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
01:04:28 Quit Roger__ ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
01:09:16 Quit daurn| (Read error: 110 (Connection timed out))
01:12:04 Quit toffe82 ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
01:12:06 Join MethoS- [0] (n=clemens@134.102.106.250)
01:18:51 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
01:22:14 Quit daurn (Read error: 110 (Connection timed out))
01:24:28 Quit MethoS- (Remote closed the connection)
01:27:31 Join MethoS- [0] (n=clemens@134.102.106.250)
01:31:50 Quit Thundercloud (Remote closed the connection)
01:32:31 Quit MethoS- (Remote closed the connection)
01:35:12 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
01:35:52 Quit JdGordon ("Leaving.")
01:38:48 Join daurn| [0] (n=daurnima@freenode/staff/daurnimator)
01:48:15 Quit tvelocity[a] ("Αποχώρησε")
01:48:58 Join mc2739 [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
01:52:45 Quit matsl (Read error: 110 (Connection timed out))
01:54:11 Quit daurn (Read error: 110 (Connection timed out))
02:00
02:04:06 Join AsaelReiter [0] (n=5f56416c@giant.haxx.se)
02:05:39 Join xavieran [0] (n=xavieran@ppp121-44-212-2.lns10.mel4.internode.on.net)
02:11:07 Quit AsaelReiter ("CGI:IRC")
02:14:58 Quit stripwax (Read error: 104 (Connection reset by peer))
02:15:25Hillshumsaratoga: ping
02:15:40saratogaHillshum: pong
02:15:56HillshumHave you uploaded any more?
02:16:10saratogahttp://duke.edu/~mgg6/rockbox/files2/
02:16:29saratogacuriously my quota seems to have been increased
02:16:37saratogathough let me know if you can't get them for some reason
02:17:22saratogaI am very interested in seeing updated benchmarks for PP, gigabeat, AMS, or TCC
02:17:55 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
02:17:57HillshumI guess the Clip isn't the best test platform due to the crashing...
02:18:22*Hillshum compiles an e200 build
02:19:00saratogacrashing is a non-issue for test_codec since it doesn't use the playback engine at all
02:19:07saratogainstead it simply won't run
02:19:15saratogaunless its been updated recently
02:22:20Hillshumahh.
02:22:34saratogayeah that hasn't been fixed
02:22:46saratogaso test codec won't work on the clip for files much bigger then a few hundred KB
02:23:18HillshumNow, why did my kernel cpu usage go up to about 20% while compiling (and stay there)?
02:23:37saratogahuh actually even the e200v2 is going to have problems since its only got 8MB
02:24:15saratogaZagor: link is above
02:25:56tmztis there something that needs tested on fuze? (v1)
02:26:14Torneyup, ipodvideo has an external mux controlled by GPIOs which routes various signals into one of the ADCs we don't currently read
02:26:21Torneincluding the voltage on D+ and D- of usb
02:26:23Hillshumtmzt: microSD
02:26:34tmztfrom svn or patches?
02:26:43Tornethe details f how it usses that to detect the charger are not 100% clear but it's obviously how it works from looking at the diag mode code
02:27:14Tornei shall have to rip the diagmode apart a little more another day and add the extra adc channels to rockbox
02:28:06CIA-43New commit by 03kugel (r22863): e200v2/Fuze: Improve scrollwheel a little bit by assuming the previous wheel value if no new value was found (which happens if a single wheel value ...
02:28:40 Join chandoo [0] (n=chandoo@ool-4353b978.dyn.optonline.net)
02:29:09saratogaHillshum: probably files bigger then 5MB won't work on the e200v2 without improvements to test_codec.c
02:29:28HillshumHmm.
02:29:45Tornegevaerts: see above, think i've found how to detect the charger on ipod :)
02:30:26saratogai don't think it would be that hard to fix (basically just have read_filebuf and request_buffer check if they're about to run out of buffer space, and if so get the current time, rebuffer, and then subtract the amount of time needed to rebuffer from the total time)
02:30:56 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
02:31:07HillshumI'm not about to fix it personally...
02:31:17saratogayes i can do it eventually but not tonight
02:31:24saratogain the mean time it would be interesting to see benchmarks on larger memory targets, or to have more codec samples from other formats
02:31:41saratogawe have nothing for AC3, Cook, ATRAC, and ALAC for instance
02:33:32saratogaalso AAC-HE and SHN
02:33:33kugelbtw, are there more containers for atrac?
02:33:41saratogayes also some Sony one
02:35:54tmztALAC can be produced with iTunes
02:36:26*kugel is bitter that the hold button doesn't work with bertrik's dbop rework
02:37:26***Saving seen data "./dancer.seen"
02:37:53saratogaproduce some then
02:55:02 Quit BlakeJohnson86 (Remote closed the connection)
02:56:52 Join BlakeJohnson86 [0] (n=bjohnson@c-24-118-162-123.hsd1.mn.comcast.net)
02:57:43 Quit Bob_C ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
02:58:54 Join arohtar [0] (n=faemir@78.33.109.163)
02:59:57 Quit jgarvey ("Leaving")
03:00
03:05:34 Quit chandoo ("Leaving")
03:06:03 Quit faemir (Read error: 60 (Operation timed out))
03:06:22 Quit saratoga ("Page closed")
03:36:04 Quit _lifeless (Read error: 148 (No route to host))
03:38:01 Quit kugel (Remote closed the connection)
03:51:42 Quit amiconn (Nick collision from services.)
03:51:45 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
03:52:06 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
03:53:50 Quit pixelma (Nick collision from services.)
03:53:52 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
03:54:11 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
04:00
04:02:45 Quit panni_ ("( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de )")
04:07:04 Quit TheSeven (Nick collision from services.)
04:07:21 Join The_Seven [0] (n=theseven@dslb-084-056-142-155.pools.arcor-ip.net)
04:07:33 Nick The_Seven is now known as TheSeven (n=theseven@dslb-084-056-142-155.pools.arcor-ip.net)
04:11:21 Quit Hillshum ("Ex-Chat")
04:13:43 Join themifflin [0] (n=themiffl@24-231-228-123.dhcp.aldl.mi.charter.com)
04:19:32 Quit themifflin ()
04:29:04 Quit Rondom (Nick collision from services.)
04:29:15 Join Rondom [0] (n=Rondom@dslb-084-057-154-162.pools.arcor-ip.net)
04:37:29***Saving seen data "./dancer.seen"
04:49:31 Quit mc2739 ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
04:50:11 Join mc2739_ [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
04:53:18 Nick mc2739_ is now known as mc2739 (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
05:00
05:11:13 Quit Llorean ("Leaving.")
05:15:22 Join Llorean [0] (n=DarkkOne@adsl-99-4-146-40.dsl.hstntx.sbcglobal.net)
05:20:21 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
05:41:47 Quit Riku (Read error: 104 (Connection reset by peer))
05:42:52 Quit Horscht ("Verlassend")
05:54:39CIA-43New commit by 03FlynDice (r22864): AMS Sansa: Replace another simple delay with a more intelligent MCI FIFO status check.
05:57:23 Join killan_ [0] (n=nnscript@c-0efa70d5.06-397-67626721.cust.bredbandsbolaget.se)
05:58:29 Quit killan (Read error: 54 (Connection reset by peer))
06:00
06:03:03 Quit mc2739 ("Leaving")
06:04:20 Join Lss [0] (n=Lss@cm46.delta91.maxonline.com.sg)
06:04:36 Join mc2739 [0] (n=mc2739@67.10.237.133)
06:05:13 Join mc2739_ [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
06:05:44 Part mc2739_ ("Leaving")
06:10:59 Join mc2739_ [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
06:12:37 Quit mc2739_ ("Leaving")
06:13:09 Quit mc2739 ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
06:13:29 Join mc2739 [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
06:37:33***Saving seen data "./dancer.seen"
07:00
07:00:16DEBUGReceived signal 15 (SIGTERM), terminating (snapshot: fplrun.c line 385)
07:00:16***Cleanup
07:00:16***Saving seen data "./dancer.seen"
07:00:16***Exit
07:04:05***Started Dancer V4.16
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05DEBUGgethostbyname(2) failed for irc.freenode.net (Success) (snapshot: netstuff.c line 99)
07:04:05***Unable to connect to irc.freenode.net on port 6667 (tried 8 times)
07:04:05***Cleanup
07:04:05***Cleanup
07:04:05***No seen item changed, no save performed.
07:04:05***Exit
07:05:45***Started Dancer V4.16
07:05:45***Connected to irc.freenode.net on port 6667
07:05:45***Logfile for #rockbox started
07:05:51***Server message 501: 'logbot :Unknown MODE flag'
07:05:51Mode"logbot :+i" by logbot
07:05:52 Join logbot [0] (i=rockbox@giant.haxx.se)
07:05:52 Join daurnimator [0] (i=daurnima@freenode/staff/daurnimator)
07:05:52 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
07:05:52 Join mc2739 [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
07:05:52 Join Lss [0] (n=Lss@cm46.delta91.maxonline.com.sg)
07:05:52 Join killan_ [0] (n=nnscript@c-0efa70d5.06-397-67626721.cust.bredbandsbolaget.se)
07:05:52 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
07:05:52 Join Llorean [0] (n=DarkkOne@rockbox/user/Llorean)
07:05:52 Join Rondom [0] (n=Rondom@dslb-084-057-154-162.pools.arcor-ip.net)
07:05:52 Join TheSeven [0] (n=theseven@dslb-084-056-142-155.pools.arcor-ip.net)
07:05:52 Join pixelma [0] (i=quassel@rockbox/staff/pixelma)
07:05:52 Join amiconn [0] (i=quassel@rockbox/developer/amiconn)
07:05:52 Join arohtar [0] (n=faemir@78.33.109.163)
07:05:52 Join BlakeJohnson86 [0] (n=bjohnson@c-24-118-162-123.hsd1.mn.comcast.net)
07:05:52 Join xavieran [0] (n=xavieran@ppp121-44-212-2.lns10.mel4.internode.on.net)
07:05:52 Join daurn| [0] (n=daurnima@freenode/staff/daurnimator)
07:05:52 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
07:05:52 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
07:05:52 Join liar|netbook [0] (n=liar@83.175.83.185)
07:05:52 Join maraz_ [0] (i=maraz@xob.kapsi.fi)
07:05:52 Join GodEater [0] (n=bibble@rockbox/staff/GodEater)
07:05:52 Join sbhsu [0] (n=a6530466@Zion.dorm.au.edu.tw)
07:05:52 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
07:05:52 Join sjh [0] (n=sjh@svana.org)
07:05:52 Join FlynDice [0] (n=FlynDice@c-24-19-225-90.hsd1.wa.comcast.net)
07:05:52 Join parafin [0] (i=parafin@paraf.in)
07:05:52 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
07:05:52 Join thegeek [0] (n=nnscript@s168c.studby.ntnu.no)
07:05:52 Join shodanX [0] (n=shodanX@jazz.informatik.uni-erlangen.de)
07:05:52 Join polobricolo [0] (n=paul@AGrenoble-257-1-118-251.w90-27.abo.wanadoo.fr)
07:05:52 Join tomers [0] (n=IceChat7@bzq-84-109-85-100.red.bezeqint.net)
07:05:52 Join antil33t [0] (n=Mudkips@119.224.12.185)
07:05:52 Join AndyI [0] (n=pasha_in@212.14.205.32)
07:05:52 Join FOAD [0] (n=dok@dinah.blub.net)
07:05:52 Join avacore [0] (i=nobody@1008ds1-rdo.0.fullrate.dk)
07:05:52 Join DaveDavenport [0] (n=qball@ipd50a4125.speed.planet.nl)
07:05:52 Join dionoea [0] (n=dionoea@videolan/developer/dionoea)
07:05:52 Join howell [0] (n=howell@rrcs-97-77-48-15.sw.biz.rr.com)
07:05:52 Join AlexP_ [0] (i=86ceaf37@rockbox/staff/AlexP)
07:05:52 Join JackWinter [0] (n=jack@vodsl-10103.vo.lu)
07:05:52 Join Rob2223 [0] (n=Miranda@p4FDCCFB4.dip.t-dialin.net)
07:05:52 Join Topy [0] (i=Topy44@f048075051.adsl.alicedsl.de)
07:05:52 Join rvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
07:05:52 Join aidy [0] (n=aidy@216.94.9.114)
07:05:52 Join z35 [0] (n=z35@ool-45717f0e.dyn.optonline.net)
07:05:52 Join Zarggg [0] (n=zarggg@65-78-69-194.c3-0.eas-ubr6.atw-eas.pa.cable.rcn.com)
07:05:52 Join timc [0] (n=aoeu@119.109.105.23)
07:05:52 Join gtkspert_ [0] (n=gtkspert@124-169-3-222.dyn.iinet.net.au)
07:05:52 Join mikroflops [0] (n=yogurt@90-231-195-226-no112.tbcn.telia.com)
07:05:52 Join evilnick_home [0] (n=evilnick@ool-4571af51.dyn.optonline.net)
07:05:52 Join HBK [0] (n=hbk@rrcs-97-77-51-170.sw.biz.rr.com)
07:05:52 Join hd [0] (i=jd@Wikipedia/HellDragon)
07:05:52 Join Sajber^ [0] (n=Sajber@h-142-237.A213.priv.bahnhof.se)
07:05:52 Join bubsy [0] (n=bubsy@94.139.72.137)
07:05:52 Join linuxguy3 [0] (n=timj@adsl-68-253-209-176.dsl.emhril.ameritech.net)
07:05:52 Join redfox [0] (n=redfox2@ns351996.ovh.net)
07:05:52 Join shadearg [0] (i=arg@ipv4.panoptix.net)
07:05:52 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
07:05:52 Join SUSaiyan [0] (n=SUSaiyan@cc84863-b.zwoll1.ov.home.nl)
07:05:52 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
07:05:52 Join Shaid [0] (i=adam@174.143.241.229)
07:05:52 Join krazykit [0] (n=kkit@c-24-218-166-241.hsd1.ma.comcast.net)
07:05:52 Join Torne [0] (i=torne@81.171.46.118)
07:05:52 Join jvd [0] (n=syscrash@poipu/developer/syscrash)
07:05:52 Join Beta2K [0] (n=beta@d24-36-68-97.home1.cgocable.net)
07:05:52 Join goffa [0] (n=goffa@70.33.8.114)
07:05:52 Join niekie [0] (i=quasselc@dreamworld.bergnetworks.com)
07:05:52 Join DirtySkeMe [0] (n=Adium@nor75-19-82-244-50-203.fbx.proxad.net)
07:05:52 Join n17ikh [0] (n=n17ikh@host-69-59-126-212.nctv.com)
07:05:52 Join Unhelpful [0] (n=quassel@rockbox/developer/Unhelpful)
07:05:52 Join jfc [0] (n=john@dpc6682208002.direcpc.com)
07:05:52 Join ch4os [0] (n=ch4os@gentoo/user/ch4os)
07:05:52 Join alexbobp [0] (n=alex@66.112.249.119)
07:05:52 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
07:05:52 Join janne_oksanen [0] (n=janne@130.234.191.99)
07:05:52 Join rasher [50] (n=rasher@rockbox/developer/rasher)
07:05:52 Join @ChanServ [0] (ChanServ@services.)
07:05:52 Join jordan` [0] (i=gromit@78.235.252.137)
07:05:52 Join kadoban [0] (n=mud@cpe-24-93-17-195.rochester.res.rr.com)
07:05:52 Join Dhraakellian [0] (n=ntryon@cpe-72-226-197-191.rochester.res.rr.com)
07:05:52 Join shaggy-h [0] (n=kiwi@87.74.127.193)
07:05:52 Join Hadaka [0] (n=naked@naked.iki.fi)
07:05:52 Join AB3JU [0] (n=dz@alt.dissonance.nl)
07:05:52 Join lyngaas [0] (n=staale@19.81-167-149.customer.lyse.net)
07:05:52 Join DaCapn [0] (i=dacapn@using.your.wireless-inter.net)
07:05:52 Join rwong [0] (n=ricky@www.roflwaffle.com)
07:05:52 Join meermanr [0] (n=meermanr@robmeerman.co.uk)
07:05:52 Join J-23 [0] (n=zelazko@unix.net.pl)
07:05:52 Join cg_ [0] (n=cromos@cable-kmi-fe71de00-186.dhcp.inet.fi)
07:05:52 Join aevin [0] (i=eivindsy@unaffiliated/aevin)
07:05:52 Join Slasheri [0] (i=miipekk@rockbox/developer/Slasheri)
07:05:52 Join elcan [0] (i=user36@64.32.20.56)
07:05:52 Join sinthetek [0] (n=sinthete@cpe-075-183-051-184.triad.res.rr.com)
07:05:52 Join pcc1 [0] (n=peter@master.pcc.me.uk)
07:05:52 Join preglow [0] (i=thomj@tvilling2.pvv.ntnu.no)
07:05:52 Join lostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
07:05:52 Join CIA-43 [0] (n=CIA@208.69.182.149)
07:05:52 Join zu [0] (n=zu@bucketheaded.eu)
07:05:52 Join markun [50] (n=markun@rockbox/developer/markun)
07:05:52 Join scorche|sh [50] (n=scorche@rockbox/administrator/scorche)
07:05:52 Join jon-kha [0] (i=jon-kha@83.150.91.127)
07:05:52 Join blithe [0] (n=blithe@72.14.176.144)
07:05:52 Join tha [0] (i=1038@130.83.160.206)
07:05:52 Join tarbo [0] (n=me@unaffiliated/tarbo)
07:05:52 Join r00s [0] (n=ru@zentrale.profitables.biz)
07:05:52 Join pjm0616 [0] (n=user@61.250.113.98)
07:05:52 Join ps-auxw [0] (n=arneb@dyn37.ps-auxw.de)
07:05:52 Join w1ll14m|_ [0] (n=nobody@84.104.81.17)
07:05:52 Join ThomasAH [0] (n=thomas@aktaia.intevation.org)
07:05:52 Join Galois [0] (i=djao@efnet.math.uwaterloo.ca)
07:05:52 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
07:05:52 Join yosafbridge [0] (n=yosafbri@ludios.net)
07:05:52 Join rphillips [0] (n=rphillip@66-90-184-91.dyn.grandenetworks.net)
07:05:52 Join Tuplanolla [0] (n=jani@unaffiliated/tuplanolla)
07:05:52 Join Res1 [0] (n=Res@user-0c6s6ir.cable.mindspring.com)
07:05:52 Join bekabug [0] (n=bekabug@adsl-068-209-234-142.sip.tys.bellsouth.net)
07:05:52 Join Overand [0] (i=overand@crappy.domain.name)
07:05:52 Join crwl [0] (n=crwlll@a91-156-100-168.elisa-laajakaista.fi)
07:05:52 Join courtc [0] (n=court@unaffiliated/courtc)
07:05:52 Join fxb__ [0] (n=felixbru@h1252615.stratoserver.net)
07:05:52 Join bzed [0] (n=bzed@devel.recluse.de)
07:05:52 Join Kohlrabi [0] (n=Kohlrabi@frustrum.nosebud.de)
07:05:52 Join crashd [0] (i=foobar@195.62.28.70)
07:05:52 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
07:05:52 Join _Tristan [0] (n=Tristan@i.dont.want.to.die.virgin.net.in)
07:05:52 Join tmzt [0] (n=tmzt@adsl-99-52-65-233.dsl.akrnoh.sbcglobal.net)
07:05:52 Join feisar- [0] (n=jljhook@irkki.fi)
07:05:52 Join Zambezi [0] (i=Zulu@80.67.9.2)
07:05:52 Join fish_ [0] (n=fish@freigeist.org)
07:05:52 Join ej0rge [0] (n=alhaz@alhaz.fttp.xmission.com)
07:05:52 Join jds2001 [0] (n=jds2001@fedora/jds2001)
07:05:59CtcpVersion from freenode-connect!freenode@freenode/bot/connect
07:05:59***Server message 477: 'logbot #rockbox :[freenode-info] channel flooding and no channel staff around to help? please check with freenode support: http://freenode.net/faq.shtml#gettinghelp'
07:10:15 Quit liar|netbook (Remote closed the connection)
07:12:08 Quit mc2739 (Remote closed the connection)
07:13:54 Quit daurn| (Read error: 110 (Connection timed out))
07:15:43 Join mc2739 [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
07:17:28 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
07:22:06 Quit daurn (Read error: 110 (Connection timed out))
07:25:22CIA-43New commit by 03tomers (r22865): FS #9157 - Tidy up some MAS35XX constants
07:28:32 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
07:36:17pixelmaLlorean: the patch is not made for the one with a broken Rec key...
07:40:22CIA-43New commit by 03tomers (r22866): FS #10226 - isp1583.c code cosmetics
07:42:05pixelmatomers: are you sure about the MAS_D? There are two supported MAS "models"
07:43:04tomerspixelma: Please explain... Do you mean replacing #else with #elif ?
07:43:59pixelmayes, I'm always curious if someone who I think does not own the hardware changes something in this target specific code
07:44:46pixelmaunfortunately I can't tell myself if it's a change for the better
07:45:46tomersYou are right regarding not owning the hw. I once stumbled upon this code (my first try to fix something) without realise it has nothing to do with the e200. Nevertheless I thought it should be fixed.
07:46:40tomersI've looked at mas35xx.c, and it seems there are only three brands of MAS_D: MAS3507D, MAS3539D and MAS2587D
07:48:31tomersI know it is a bit dangerous... This is a special case I decided to risk - I am not going to do that regulary. The build results are all green, Does it tells it is OK?
07:51:09 Join chandoo [0] (n=chandoo@ool-4353b978.dyn.optonline.net)
07:56:58 Join flydutch [0] (n=flydutch@host77-167-dynamic.15-87-r.retail.telecomitalia.it)
07:57:47 Quit GeekShadow ("The cake is a lie !")
08:00
08:01:49 Quit evilnick_home (Read error: 104 (Connection reset by peer))
08:02:39 Join evilnick_home [0] (n=evilnick@ool-4571af51.dyn.optonline.net)
08:07:17 Join mt [0] (n=MTee@rockbox/developer/mt)
08:11:25pixelmatomers: there's also a MAS_F if I remember correctly and I just wondered if the #else was there to catch both but I better leave this to someone else... (e.g. amiconn)
08:19:04amiconntomers: There are 3 different MAS types used in rockbox targets, but they're not all MAS_D. The 3539 and 3587 are MAS_F (MAS3539F and MAS3587F)
08:19:21tomersamiconn: Thanks. I'll revert this
08:19:23amiconnThe commit looks okay though
08:20:14tomersI won't revert. Just checked and indeed it looks ok
08:20:26 Join ender` [0] (i=krneki@foo.eternallybored.org)
08:21:13amiconnI was confused a bit about the MAS_REG_KBASS and MAS_REG_KTREBLE for the 3507D, but that's okay as well
08:22:07tomersLook in the FS item for the specs that are attached...
08:22:59*amiconn has all MAS manuals since he has targets with all three MAS types
08:26:29amiconnIn general, if you're changing something low-level for target(s) you don't own, it's usually a good idea to ask for testing
08:27:41 Join shaggy-w [0] (n=kiwi@host-87-74-127-193.dslgb.com)
08:29:01amiconnpixelma: The various #else's were there to catch the 3507D, assuming we'll never see another target with a different MAS
08:29:40 Join bertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl)
08:30:44Lloreanpixelma: Well, that's what the thread is about, and Petur called it "a fix" in that thread and the only "problem" reported is the broken Rec key, so I don't understand.
08:30:50amiconnImo this assumption will hold, although other MAS types exist (most of which probably don't fall in the 'D' but in the 'F' group, e.g. the MAS3539F is part of the MAS35x9F family, where the x can be several different digits)
08:30:53LloreanWhat's it supposed to fix, in that thread, other than the problem the thread is about?
08:30:58 Quit shaggy-h (Read error: 110 (Connection timed out))
08:30:58 Nick shaggy-w is now known as shaggy-h (n=kiwi@host-87-74-127-193.dslgb.com)
08:31:07 Part safetydan ("Leaving.")
08:33:59 Join esperegu [0] (n=quassel@145.116.11.103)
08:34:39 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
08:34:52bertrikFlynDice, in r22862 you changed sdhc to sd_v2, I think it's not automatically the case that v2 cards are also high capacity
08:35:59bertrikNot sure if it is wrong yet (it's been a while since I looked at that), but I just like to check
08:38:24tomersamiconn: I'll take your advice regarding the testing with other users. Regarding the #elif, I've double checked it don't hurt anything, and prefered to have a more strict testing, so newer MAS types will have to strictly define their values
08:40:26 Quit evilnick_home (Read error: 104 (Connection reset by peer))
08:40:45 Join evilnick_home [0] (n=evilnick@ool-4571af51.dyn.optonline.net)
08:41:56tomersOne more thing about testings - the FS was there for a very long time... I agree I should have actively ask people for testing, though :-)
08:44:06 Join Rob2222 [0] (n=Miranda@p4FDCCD88.dip.t-dialin.net)
08:49:34 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
08:54:55 Quit Topy (Read error: 110 (Connection timed out))
08:55:55 Quit daurnimator (Read error: 110 (Connection timed out))
08:56:35 Join Thundercloud [0] (i=thunderc@persistence.flat.devzero.co.uk)
08:58:09 Quit chandoo ("Leaving")
08:58:10 Quit arohtar (Client Quit)
08:58:50 Join swiftkick [0] (n=Miranda@38.106.186.56)
09:00
09:01:58 Join Topy44 [0] (n=Topy44@g228147214.adsl.alicedsl.de)
09:02:30 Quit Rob2223 (Read error: 110 (Connection timed out))
09:02:54 Quit ender` (" Computers work fine as long as users aren't allowed anywhere near them.")
09:05:21pixelmaLlorean: my guess was that petur thought the original poster just concludes this because he can't start a recording with it anymore - petur or the original poster are the ones who to speak to. I just wanted to point out that there was a change in the code which might lead people to think so (and which this patch is supposed to fix)
09:05:48***Saving seen data "./dancer.seen"
09:06:21LloreanGotcha. Yeah, it makes more sense after reading your forum post
09:06:48 Join LinusN [0] (n=linus@gateway/web/cgi-irc/labb.contactor.se/x-yecivhelkjqzjiff)
09:10:59 Quit AlexP_ ("Page closed")
09:11:03 Join petur [50] (n=petur@rockbox/developer/petur)
09:11:04 Join ender` [0] (i=krneki@foo.eternallybored.org)
09:12:49pixelmatomers: RockboxTesting in the wiki for a list of IRC nicks who you could ask for testing on a specific target
09:20:46swiftkickDear Diary,
09:20:46swiftkickLast year I replaced all the windows in my house with those expensive double pane energy-efficient kind. Then, this week I got a call from the contractor who installed them. He was complaining that the work had been completed a whole year ago and I still hadn't paid for them. Now just because I'm blonde doesn't mean that I am automatically stupid.
09:21:09swiftkickSo, I told him just what his fast talking Sales guy had told me last year.... Namely, that in ONE YEAR these windows would pay for themselves! Helllooooo? It's been a year! There was only silence at the other end of the line,so I finally just hung up. He didn't call back. Bet he felt stupid.
09:21:26swiftkickoops sorry guys
09:21:30swiftkickargh
09:21:34 Part swiftkick
09:23:37 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:24:02 Quit bertrik ("De groeten")
09:27:36 Join arkascha [0] (n=arkascha@195.227.110.226)
09:27:57 Part arkascha ("Konversation terminated!")
09:30:09 Quit JdGordon ("Leaving.")
09:30:52 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
09:42:26 Join DerPapst [0] (n=DerPapst@p4FE8FF56.dip.t-dialin.net)
09:44:49 Quit Thundercloud (Remote closed the connection)
09:48:15 Join robin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
10:00
10:22:15CIA-43New commit by 03zagor (r22867): Added necessary styles for wiki search.
10:27:41 Part linuxstb ("Leaving")
10:28:16Zagorcodec test files are now available here: http://download.rockbox.org/test_files/
10:39:53 Quit TheSeven ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
10:47:27 Join TheSeven [0] (n=theseven@dslb-084-056-142-155.pools.arcor-ip.net)
11:00
11:00:50 Join _lifeless [0] (n=lifeless@90.151.45.208)
11:05:52***Saving seen data "./dancer.seen"
11:18:44robin0800Zagor: Looking at your test files what is used to decode Mp3 Vbr ?
11:23:21Zagorrobin0800: saratoga made these files. from the filename I assume the mp3 files are all cbr
11:23:49 Join matsl [0] (n=matsl@91.191.140.126)
11:29:28LloreanZagor: I know there's several lossless files, is there some reason the source .WAV isn't there too (I guess we don't really need to benchmark the .wav "codec"?
11:31:31Zagorno. I just put up the files saratoga made.
11:33:44Zagorthe wiki page links to a flac file as source
11:35:41LloreanI don't know how much sense .wav would make anyway (at least for test_codec). I guess it would serve to show the "maximum" speed though.
11:38:49 Join howell7 [0] (n=howell@rrcs-97-77-48-15.sw.biz.rr.com)
11:42:33 Quit howell (Read error: 110 (Connection timed out))
11:50:20 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net)
11:54:37 Quit TheSeven ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
12:00
12:00:12 Join TheSeven [0] (n=theseven@dslb-084-056-142-155.pools.arcor-ip.net)
12:11:27 Quit rvvs89 (Remote closed the connection)
12:11:34 Join rvvs89 [0] (n=ivo@bright-snat.ucc.asn.au)
12:22:43 Join notlistening [0] (n=tom@94-195-105-95.zone9.bethere.co.uk)
12:23:38notlisteningHi, I just tried to post a private message on the forums and the verfication box is totally unreadable, even for a fully sighted person is this a know issue
12:25:39LloreanDid you try clicking "request another image" a few times?
12:26:16 Join einhirn [0] (n=Miranda@bsod.rz.tu-clausthal.de)
12:26:39*gevaerts encourages any feature that keeps his PM inbox empty ;)
12:28:28LloreanI've found a quite good solution to that, actually. My inbox is currently "1728% full" since it went from unlimited to the normal-user limit. :)
12:30:07LloreanI do notice that the registration captcha seems to occasionally generate "bad" ones where you several of the characters aren't there at all.
12:30:47TheSevenyep, also got such a thing back when i was registering
12:40:46 Quit _lifeless (Read error: 148 (No route to host))
12:41:26 Join uflops [0] (n=yogurt@90-231-195-226-no112.tbcn.telia.com)
12:45:35pixelmanotlistening: there's also a button to let it read out for you
12:46:09notlisteningOkay i have to go back and see again because i am being more blnd than usual
12:47:50notlisteninghttp://forums.rockbox.org/index.php?action=useremail;u=24001
12:48:12notlisteningnpw as far as i can see there is no regenerate button or listen to options
12:50:16LloreanYou're trying to send an email, not a PM. Just reload the page.
12:50:39LloreanI do suspect it's broken though, as I'm getting the same code every time.
12:50:41Tornepoll: has anyone ever plugged their dap into a usb port without enough power to run it and been annoyed that you can't get it to just talk usb without charging? :)
12:51:18LloreanTorne: Which DAP does this?
12:51:47Torneall rockbox targets with swusb
12:51:56Tornethe descriptors in usb_core.c always request 500mA
12:52:29Torneso properly behaving USB stacks will refuse to talk to us at all unless 500mA is available
12:53:20Torne(this is regardless of whether we actually support usb charging on that target, or whether we have a setting for whether usb charging is allowed)
12:54:35Tornei'm looking at fixing up the ipodvideo (+other) usb charging support, and am wondering if it's worth the effort of making the usb charging enable setting actually change the descriptor on targets where it's possible for us to actually regulate our current usage.
12:54:57Torne(not entirely sure which targets this is, but ipodvideo is one)
12:55:01gevaertsif we have the setting, I think it's a good idea
12:55:01 Quit mikroflops (Read error: 110 (Connection timed out))
12:55:50Tornethe setting is there already on some targets
12:56:13Tornegevaerts: i found vaguely how apple detect chargers, btw
12:56:21gevaertsI saw, yes
12:56:51gevaertsof course now you have to write code to make use of it :)
12:56:57Torneso i'm looking now at implementing spec compliant behaviour for actual USB hosts first
12:57:16Tornethen will do the A2D stuff for charger detection later once i've read a bit more of diagmode.ipod :)
12:57:35Torne(we might need some user help to see what values different chargers give)
12:58:49Tornei figured since i seem to be one of the primary people arguing *against* committing the patch for fs#8802 in its present state that i should see about writing a better one ;)
12:59:01LloreanDidn't we used to have charger detection?
12:59:20Torneon ipod? not htat i know of
12:59:24LloreanYes, on iPod
12:59:27LloreanLook back a few years.
12:59:33LloreanLong before we had swusb.
12:59:33Tornehmm
13:00
13:00:00LloreanIt got broken by a "fix" later on, but for a period it would reboot to disk mode if you attached to a computer, or simply charge if you attached a dedicated charger.
13:00:09pixelmayes, it killed the automatic reboot into the OF on USB plug in some builds
13:00:20gevaertsI think we had some sort of activity detection
13:00:25TorneOh.. hm.
13:00:40*gevaerts isn't sure. He wasn't there
13:00:42Tornesuggestion where to look? ;)
13:00:58LloreanSorry, couldn't say.
13:01:04pixelmain cases where it should reboot (USB data connection)
13:01:05LloreanMy first guess would be "commits by linuxstb or preglow"
13:01:21LloreanIt was exclusively an iPod thing, and for quite some time they were about the only people committing ipod-specific code.
13:02:39Tornehttp://svn.rockbox.org/viewvc.cgi/trunk/firmware/target/arm/usb-fw-pp502x.c?r1=15772&r2=15774 <- looks related
13:04:13Tornei don't htink that's the same kind of detection..
13:04:28pixelmaI think it was committed by the end of 2006 as I remember having trouble with amiconn's 2nd gen Mini, he lent me during my Ondio's backlight mod
13:05:08Tornehttp://svn.rockbox.org/viewvc.cgi?view=rev&revision=12198 possibly
13:05:21LloreanTorne: I don't know if it's the same kind, I just know the behaviour it had on my Nano - if I used a dedicated charger (Griffin car charger, for example) it didn't reboot and behaved as other Rockbox players do on wall power, if I attached it to my PC it rebooted into disk mode.
13:05:35TorneLlorean: yah, but that works now also
13:05:49Tornei'm not talking about telling the difference between a host and a non-host
13:05:54***Saving seen data "./dancer.seen"
13:06:04Tornei'm specifically talking about identifying an Apple AC adapter via the adc readings from D-/D+ pins
13:06:25Tornewhich signals that it's safe to draw lots of power without negotiating with a host (apple's own usb charging "spec", before the real one was written)
13:06:47LloreanI'm confused then between "non-host" and "charger"?
13:06:55LloreanWhat provides power, is non-host, and isn't a charger?
13:07:52gevaertsa nuclear power plant
13:07:55*gevaerts runs away
13:07:58Tornea PC with a crashed usb stack
13:08:09Tornea charger that can't deliver 500mA
13:08:10Torneetc
13:09:12LloreanYes, but does that mean we're going to intentionally *not* draw 500mA from chargers that don't identify themselves "just in case"?
13:09:22Tornethat's what i'm suggesting, yes
13:09:29Tornealso known as "complying with the usb spec"
13:09:29LloreanThat seems a bad idea.
13:10:05LloreanI have about a half-dozen chargers for various devices (most of them 1000mA) with USB plugs. What's the likelihood most/any would pass your test?
13:10:17Tornedepends.
13:10:30Tornethey might implement the standard way from the new usb charging spec, possibly :)
13:10:36TheSevenTorne: not even complying with the USB spec. That requires a far lower current if there is no bus activity
13:10:37Tornewhich I *think* the ipod has the hardware to detect. but not sure.
13:10:49TorneTheSeven: er, what?
13:10:53LloreanTorne: If it's a new USB spec, could it be on a charger two years old?
13:10:59TorneLlorean: unlikely
13:11:01*TheSeven needs to look it up
13:11:09LloreanIt seems like this is a way to simply lock out a lot of chargers and confuse users.
13:11:12TorneTheSeven: You can draw 100mA as soon as you get voltage on pwr/gnd
13:11:24TorneLlorean: at the moment we don't draw 500mA from *anything* on ipod
13:11:27Tornehost or not, charger or not
13:11:35Torneso it *never* charges properlt
13:11:51LloreanWhich is also bad.
13:11:57TorneTheSeven: if you have no power supply of your own, you need to :)
13:12:14LloreanTorne: You're proposing replacing "bad" with "a different bad" though from my perspective.
13:12:37LloreanSaying "chargers are support" won't really be true if it's *only* chargers made specifically for iPods, and ones made in the last X months.
13:13:42TorneLlorean: yes, i know
13:14:04TheSeven[13:09]<Torne>TheSeven: You can draw 100mA as soon as you get voltage on pwr/gnd
13:14:06TheSevenwrong.
13:14:13LloreanI mean I doubt my cheap Energizer AA USB charger will pass the test either.
13:14:24LloreanAnd if it's only having 100mA drawn from it, it's pretty much pointless.
13:14:45TorneTheSeven: really?
13:14:56TheSevenlet me quote it from the specs...
13:15:53TheSeven500µA
13:16:21Torne..really? when do you get to draw the unit load, then?
13:16:54TheSevenas soon as there is bus activity
13:17:23Tornefun
13:17:34TheSeventhen you may draw up to 100mA, and if you need more, you must request that from the OS, which will check if enough current is available in that place
13:17:58Tornewell, that makes even the useless behaviour we have non-compliant then :)
13:18:06TheSevenjust drawing whatever you want isn't an issue at all if you're dealing with chargers, but it is, if the ipod is connected to a bus-powered USB hub
13:18:15Torneyes, i know
13:18:20Tornei tested this with my keyboard just now :)
13:18:37Torneit makes everything else on the hub stop working
13:18:40Tornewhich is unfortunate :)
13:19:25TheSeventhere are 2 possible reactions in that case: either the hub shuts off power to that port, or if it's a cheap hub, it can't do that, so the host shuts off power to the hub.
13:20:43TheSevensee page 178 (PDF page 206) of the USB2.0 specs
13:20:58Torneright.
13:21:11TorneLlorean: what behaviour do you want, then? :)
13:21:38Tornethe patch on fs#8802 adds direct control of the 100/500mA select pin on the power controller
13:21:47TheSevenhonestly, I'd go for 100mA if there is bus activity, and 500mA else
13:21:49 Join efyx_ [0] (n=efyx@lap34-1-82-225-185-146.fbx.proxad.net)
13:21:52Torneif you turn the usb charge setting on it draws 500mA, whatever you plug it into, even if you don't get permission
13:21:58Torneif you turn it off, 100mA always
13:22:10Tornedreamlayers didn't intend this to be committed as is :)
13:22:13LloreanThat's more or less what I'd like to see. A manual choice between 500 and 100, and never letting the software decide (unless there's a guaranteed method to know what can be provided)
13:22:17Tornebut it proves the GPIOs work
13:22:31TorneLlorean: for USB hosts, there *is*.
13:22:49LloreanTorne: Exactly, so in their case, the software should decide *if* the host says it can't provide as much as we want.
13:23:02LloreanIf the user says 100 and the host says "but yeah, I can do 500" it shouldn't jump up, though.
13:23:12Torneyes.
13:23:13Tornethat's fine
13:23:25Tornethe original question i asked at the stard of all this is "should we change what we ask for", though
13:23:29Tornewhich i think the answer to is yes
13:23:32LloreanBasically, the user is deciding a maximum, and what to do if there's no host to negotiate with
13:23:42Torne(we currently always ask for 500 even if we are *guaranteed* not to use it)
13:23:50Torne(which means you cna't connect to low power hubs at all)
13:24:02TheSevenUSB activity, unconfigured => 100mA, USB activity, configured, but didn't allow more current => 100mA, USB activity, configured, 500mA allowed => 500mA, no USB activity => 500mA seems most compatible to me
13:24:27LloreanIs it possible to say "draw as much as you can get" rather than capping at 500?
13:24:30TorneTheSeven: unfortunately allowing current doesn't work that way
13:24:35LloreanI thought a lot of Apple chargers are actually 1000
13:24:36TorneLlorean: Not sure.
13:24:40Torneyes, they are
13:24:43TheSevenTorne: why?
13:24:45Tornebut i don't think the power controller has that option
13:24:52LloreanAnd I *think* i've got a 2000 "USB" charger around here somewhere
13:24:54TorneTheSeven: if you ask for 500 and the host syas no then you don't get configured
13:24:55 Join gapan [0] (n=gapan@adsl210-138.lsf.forthnet.gr)
13:25:04TorneTheSeven: you can't change your mind and keep connecting, afaik
13:25:04Zagoryes, both my iphone chargers (one apple, one third party) are 1000 mA
13:25:16TorneYes
13:25:23Tornethe iphone uses the actual usb chargingn spec, afaict
13:25:58TheSevenTorne: that should work by just providing 2 configuration descriptors, one high-power and one low-power one, and letting the host choose
13:25:59Tornewhich just says "if D+/D- are shorted with 200ohm then pull as much as you like, we are a regulated charger"
13:26:02 Join daurn| [0] (n=daurnima@freenode/staff/daurnimator)
13:26:03 Quit efyx_ (Remote closed the connection)
13:26:04ZagorTheSeven: how long do you wait for bus activity before using full power?
13:26:21TorneI don't know if we have a bit on the ipodvideo to turn *off* the power controller's regulated input.
13:26:21TheSevenor you could do a soft disconnect if 500 fails, and then provide the 100 descriptor
13:26:33TorneTheSeven: ok, but i'm not about to implement that :)
13:26:36Tornethat's too much usb for me :)
13:27:34TheSevenZagor: i think you may draw full power immediately, and need to go down to 500µA again if there was no activity for 10ms or some such, need to check
13:27:37Torneso yeah. the problem is how do you decide that the thing you're connected to *isn't* a host, and that you aren't going to upset it by pulling lots
13:28:01Tornealso i't snot clear exactly which GPIOs to set to *stop* drawing power ;)
13:28:14Torneatm it always does 100 or 500; there is some way to make it stop but i'd have to fiddle to work that one out
13:30:26TheSevenZagor: yep, it is 10ms
13:30:55Zagorisn't that odd? you get to pull insane amounts of power for 10ms?
13:31:00Tornewait, you're allowed to draw full power immediately, but then have to stop after 10ms if you've not been enumerated?
13:31:09 Quit shodanX (Read error: 110 (Connection timed out))
13:31:11TheSeven[13:26]<Torne>atm it always does 100 or 500
13:31:17TheSevennano2 OFW is doing that, too
13:31:32TheSevenso it will draw 100 as soon as there is power, no matter the usb state
13:31:32TorneTheSeven: i'm not at all doubting that most or even all of the OFs do that on every player :)
13:31:58Tornethat still doesn't mean it's right..
13:32:58TheSevenZagor: as soon as there is activity, you may draw 100mA, for up to 10ms, and need to fall back to 500µA after that. Once you are configured, there's usually keepalive traffic every 3ms or such to stop that from happening
13:33:49TheSevenif it isn't, it means the host fell into some kind of standby mode, and that all devices must go back to 500µA within 10ms
13:33:55Zagorright. so when can you pull the full 1000 mA? or is apple screwing the spec like everyone else?
13:34:21gevaertsTheSeven: where do you get that 500µA requirement?
13:34:35TorneZagor: you can't have 1A unless you do the new usb charging spec.
13:34:49Tornewhich is "notice that D+ and D- are shorted together", more or less
13:34:59Torneat which point you are allowde to turn your regulator off and drink all you are given iirc
13:34:59ZagorTorne: ok
13:35:01TheSevenZagor: you can pull 500mA when you're configured, there is activity, and the OS has granted that much power to you
13:35:22TorneZagor: Apple *also* have their own way of signalling this
13:35:28Tornewhich they used before the usb charging spec was written
13:35:32Tornewhich appears to be different.
13:35:56 Join funman [0] (n=fun@rockbox/developer/funman)
13:36:01TheSevenTorne: not only that, you may try to pull more than 1000mA to make the charger apply current limiting, to adjust the voltage you get and therefore reduce heat dissipation while charging
13:36:05Torne(D+ and D- pulled up/down to vcc/gnd respectively via particular resistances, detected via ADC)
13:36:23gevaertsTheSeven: "suspended" is not the same as "no activity"
13:36:24TorneZagor: it looks like ipodvideo at least supports both
13:36:31TheSevengevaerts: [13:19]<TheSeven>see page 178 (PDF page 206) of the USB2.0 specs
13:36:42Torneso it will work with both old apple chargers, new apple chargers, or with new USB-charging-spec compliant chargers. I think.
13:36:53TheSevenyep, but no activity leads to suspended after 10ms
13:36:57gevaertsno
13:37:32Torneanyway, yes, TheSeven clearly knows the specs better than i do, i hadn't yet got to the stage of actually checking any of this
13:37:56Tornebut the fundamental question remains: should we actually follow the specs all the time, or should we try and charge from USB power sources which don't have any wya of telling us it's ok to do so?
13:38:32TheSevengevaerts: checking again, but I think I read that somewhere in chapter 9
13:38:41Tornethe details of how to handle the negotiation with hosts is good and we should really do that right, because there's not really a downside to doing so when we're connected to a host.
13:38:53gevaertsTheSeven: well, yes, but that sort of "no activity" is not the same as e.g. no UMS traffic
13:39:21gevaertsAlso, it can't happen before enumeration
13:39:37Tornei'm personally happy with spec compliant behaviour, but that's because the only usb charger i have follows the spec.
13:39:45Torne:)
13:39:52TheSeven"In order to conserve power, USB devices automatically enter the Suspended state when the device has observed no bus traffic for a specified period (refer to Chapter 7). When suspended, the USB device maintains any internal status, including its address and configuration.
13:39:54TheSevenAll devices must suspend if bus activity has not been observed for the length of time specified in Chapter 7. Attached devices must be prepared to suspend at any time they are powered, whether they have been assigned a non-default address or are configured. Bus activity may cease due to the host entering a suspend mode of its own. In addition, a USB device shall also enter the Suspended...
13:39:56TheSeven...state when the hub port it is attached to is disabled. This is referred to as selective suspend."
13:40:52gevaertsok. I was wrong about the enumeration bit...
13:41:05 Join chb [0] (n=1000@unixboard/mod/chb)
13:42:01TheSevengevaerts: yes, I was talking about bus activity, not UMS traffic, and as I said, there is some keepalive bus traffic to prevent that from happening as long as the host is up and running (and thus able to supply those 100mA)
13:42:04gevaertsWe don't actually handle suspend at all though, and I'm pretty sure that no OS suspends devices during an actual MSC connection
13:42:26gevaertsthe SOF packets are still sent, yes
13:42:42*gevaerts votes to ignore the suspend case for now
13:42:47 Quit daurn (Read error: 101 (Network is unreachable))
13:43:06TheSevengevaerts: an OS *should* release the configuration before shutting something down, but what if it e.g. crashes? :-)
13:43:45TheSevenand yes, i also vote to ignore that suspend case, as I've never seen any problems with drawing 100mA. This suspend case is mainly intended for battery-operated hosts.
13:44:22TheSeventhe question was just when to draw 100mA or 500mA, and i just stated that not even drawing 100mA all the time is really compliant because of this suspend thing.
13:45:45gevaertsI also suspect that enough devices handle suspend wrong that OSes don't rely on it
13:47:38Torneok.
13:47:47TheSevenactually, almost every device i've seen until today handles that one, and it looks like the USB IF even certifies most of them
13:47:56Torneso i don't think anyone *disagrees* that it'd be nice if we did the 100/500 switching when we had a host properly
13:48:09Torneand i can probably do that quite easily for ipod (dunno about other targets)
13:48:10 Part gapan
13:48:29Tornebeast already does it, and looks like it's probably rihgt
13:48:34gevaertsI think we should do 100/500 in a spec compliant way, and add a setting to allow getting 500 anyway if no host is detected
13:48:55TheSevengevaerts: sounds good
13:48:56gevaertsand possibly a setting to only ask for 100
13:49:04gevaertsi.e. no USB charging
13:49:04Tornegevaerts: there's already a setting for "enable usb charging"
13:49:14Tornewhat this does on various targets seems to differ though :)
13:49:19gevaertsok, "keep a setting to ask for 100" :)
13:49:55Torneso how do we do the no host thing?
13:49:56Tornetimeout?
13:50:01Tornethen go with the user's preference?
13:50:01 Join pamaury [0] (n=pamaury@140.77.26.40)
13:50:04TheSevenTorne: the 100/500 switching is done by a GPIO on the old ipods?
13:50:09CIA-43New commit by 03markun (r22868): add support for comments in ID3v2.2
13:50:13TorneTheSeven: Not sure about *old* ipods
13:50:18pamaurypcc1: hello
13:50:23Torneunless you mean "Ones we support already"
13:50:36TheSevenwith old, i mean the portalplayer ones
13:50:38FlynDicebertrik:(for the logs) RE: r22862 That is correct, not all v2 cards are necessarily high capacity but the test at that point does not determine if the card is HC but only if it is V1 or V2. If it is V2 we need to send the HCS bit along with the allowed voltage range. If it replies with the CCS bit set it is HC.
13:50:39Torneright
13:50:47Tornethe more recent portalplayer ones have a GPIO for it
13:50:55Tornethe less recent portalplayer ones i think are unknown
13:51:22TheSevenok, on the Samsung ones, it's all I2C
13:51:28Torneright
13:51:32Tornei kno wnothing about those
13:51:54TheSevenactually I'm right now trying to find out how to enable 500mA on them
13:52:09Tornethe GPIO toggling patch dreamlayers wrote is for 4g/5g, mini 2g, nano 1g
13:52:17TheSeven(as 100mA is not even sufficient to stop them from discharging in some cases)
13:52:28Torne100mA is not enough to stop the PP-based ones discharging either
13:52:40amiconnLlorean: Test fils for .wav *might* actually make sense. Not so much for plain pcm, but for adpcm and stuff like that
13:52:51TheSevenwell, for the nano, it proabably is, as long as you don't open all clock gates at a time, like ibugger does
13:53:00TorneTheSeven: well, maybe
13:53:03chbare there any voice files for the sansa v2 players - I can only find v1 voice files
13:53:06Torneit's not enough for the hard disk based pp ones )
13:53:18Tornewhich is the case i care about personally ;)
13:53:20Tornehaving an ipodvideo
13:54:13TorneIs there any value at all in trying to detect chargers via the adc then?
13:54:18Tornei guess there is
13:54:38Tornebecause even if you have a setting to say "draw 500ma from anything"..
13:54:42Torneyou could ignore tha tsetting being set to no if you detect that you are connected to a spec compliant charger (or to an apple charger)
13:55:52markundid my patch cause the red?
13:56:29markunlooks like some internal compiler error
14:00
14:11:28 Join shodanX [0] (n=shodanX@jazz.informatik.uni-erlangen.de)
14:22:01markunamiconn: do the HWCODEC targets have support for gapless playback?
14:22:17 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
14:24:59 Quit robin0800 (Remote closed the connection)
14:29:32Zagormarkun: yes, but only with specially massaged mp3 data
14:31:14pamauryhello, I'm sure I've already asked it but how with git can I sync a branch (say mtp-experimental) with another (say master) to have all recent changes applied to it ?
14:31:56 Join robin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
14:35:29markunZagor: ok, then I guess my itunes gapless patch could use some ifdefs for SWCODEC
14:36:50 Quit esperegu (Read error: 113 (No route to host))
14:37:12Zagormarkun: yes
14:39:43 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
14:40:30 Join merbanan [0] (n=banan@c-83-233-172-245.cust.bredband2.com)
14:41:15funmanpamaury: git pull?
14:42:20pamauryfunman: I remember someone telling me that there could be some problem with the history
14:42:46funmanpamaury: git pull −−rebase will reapply your local commits on top of the merged branch
14:43:06funmanproblems with history only matter when you redistribute your whole tree to other people, not when you only redistribute patches
14:44:02pamauryah ok
14:44:49funmanlocal history can always be changed (save all the patches, reset −−hard, apply all the patches again for example)
14:45:24pamauryyes you're right
14:48:07 Join esperegu [0] (n=quassel@s559081d2.adsl.wanadoo.nl)
14:52:34 Quit daurn| (Read error: 110 (Connection timed out))
15:00
15:01:10 Quit pamaury ("exit(*(int *)0 / 0);")
15:05:56***Saving seen data "./dancer.seen"
15:06:53 Join likemindead [0] (n=XXXXXXXX@ama-wldhcp.69614189.amaonline.com)
15:10:55 Part LinusN
15:10:56 Join panni_ [0] (i=hannes@ip-95-222-21-143.unitymediagroup.de)
15:19:55 Quit robin0800 (Remote closed the connection)
15:20:26 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
15:20:53 Join robin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
15:26:12 Quit stoffel (anthony.freenode.net irc.freenode.net)
15:26:12NSplitanthony.freenode.net irc.freenode.net
15:26:12 Quit robin0800 (anthony.freenode.net irc.freenode.net)
15:26:12 Quit dfkt (anthony.freenode.net irc.freenode.net)
15:26:12 Quit einhirn (anthony.freenode.net irc.freenode.net)
15:26:12 Quit rvvs89 (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Topy44 (anthony.freenode.net irc.freenode.net)
15:26:12 Quit pixelma (anthony.freenode.net irc.freenode.net)
15:26:12 Quit amiconn (anthony.freenode.net irc.freenode.net)
15:26:12 Quit parafin (anthony.freenode.net irc.freenode.net)
15:26:12 Quit daurn (anthony.freenode.net irc.freenode.net)
15:26:12 Quit JdGordon (anthony.freenode.net irc.freenode.net)
15:26:12 Quit mc2739 (anthony.freenode.net irc.freenode.net)
15:26:12 Quit maraz_ (anthony.freenode.net irc.freenode.net)
15:26:12 Quit FlynDice (anthony.freenode.net irc.freenode.net)
15:26:12 Quit sjh (anthony.freenode.net irc.freenode.net)
15:26:12 Quit sbhsu (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Bagder (anthony.freenode.net irc.freenode.net)
15:26:12 Quit GodEater (anthony.freenode.net irc.freenode.net)
15:26:12 Quit merbanan (anthony.freenode.net irc.freenode.net)
15:26:12 Quit chb (anthony.freenode.net irc.freenode.net)
15:26:12 Quit gevaerts (anthony.freenode.net irc.freenode.net)
15:26:12 Quit antil33t (anthony.freenode.net irc.freenode.net)
15:26:12 Quit tomers (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Sajber^ (anthony.freenode.net irc.freenode.net)
15:26:12 Quit niekie (anthony.freenode.net irc.freenode.net)
15:26:12 Quit AB3JU (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Hadaka (anthony.freenode.net irc.freenode.net)
15:26:12 Quit esperegu (anthony.freenode.net irc.freenode.net)
15:26:12 Quit uflops (anthony.freenode.net irc.freenode.net)
15:26:12 Quit TheSeven (anthony.freenode.net irc.freenode.net)
15:26:12 Quit DerPapst (anthony.freenode.net irc.freenode.net)
15:26:12 Quit shaggy-h (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Rondom (anthony.freenode.net irc.freenode.net)
15:26:12 Quit AndyI (anthony.freenode.net irc.freenode.net)
15:26:12 Quit DaveDavenport (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Galois (anthony.freenode.net irc.freenode.net)
15:26:12 Quit meermanr (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Tuplanolla (anthony.freenode.net irc.freenode.net)
15:26:12 Quit rwong (anthony.freenode.net irc.freenode.net)
15:26:12 Quit J-23 (anthony.freenode.net irc.freenode.net)
15:26:12 Quit lyngaas (anthony.freenode.net irc.freenode.net)
15:26:12 Quit rasher (anthony.freenode.net irc.freenode.net)
15:26:12 Quit pcc1 (anthony.freenode.net irc.freenode.net)
15:26:12 Quit r00s (anthony.freenode.net irc.freenode.net)
15:26:12 Quit LinusN (anthony.freenode.net irc.freenode.net)
15:26:12 Quit avacore (anthony.freenode.net irc.freenode.net)
15:26:12 Quit hd (anthony.freenode.net irc.freenode.net)
15:26:12 Quit krazykit (anthony.freenode.net irc.freenode.net)
15:26:12 Quit Beta2K (anthony.freenode.net irc.freenode.net)
15:26:12 Quit DirtySkeMe (anthony.freenode.net irc.freenode.net)
15:26:12 Quit cg_ (anthony.freenode.net irc.freenode.net)
15:26:12 Quit preglow (anthony.freenode.net irc.freenode.net)
15:26:12 Quit lostlogic (anthony.freenode.net irc.freenode.net)
15:26:12 Quit sinthetek (anthony.freenode.net irc.freenode.net)
15:26:12 Quit zu (anthony.freenode.net irc.freenode.net)
15:26:12 Quit yosafbridge (anthony.freenode.net irc.freenode.net)
15:26:12 Quit rphillips (anthony.freenode.net irc.freenode.net)
15:26:12 Quit DaCapn (anthony.freenode.net irc.freenode.net)
15:26:12 Quit CIA-43 (anthony.freenode.net irc.freenode.net)
15:26:20 Quit flydutch (anthony.freenode.net irc.freenode.net)
15:26:20 Quit polobricolo (anthony.freenode.net irc.freenode.net)
15:26:20 Quit bubsy (anthony.freenode.net irc.freenode.net)
15:26:20 Quit goffa (anthony.freenode.net irc.freenode.net)
15:26:20 Quit ThomasAH (anthony.freenode.net irc.freenode.net)
15:26:20 Quit ps-auxw (anthony.freenode.net irc.freenode.net)
15:26:20 Quit markun (anthony.freenode.net irc.freenode.net)
15:26:20 Quit funman (anthony.freenode.net irc.freenode.net)
15:26:20 Quit matsl (anthony.freenode.net irc.freenode.net)
15:26:20 Quit dionoea (anthony.freenode.net irc.freenode.net)
15:26:20 Quit aidy (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Shaid (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Torne (anthony.freenode.net irc.freenode.net)
15:26:20 Quit alexbobp (anthony.freenode.net irc.freenode.net)
15:26:20 Quit janne_oksanen (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Slasheri (anthony.freenode.net irc.freenode.net)
15:26:20 Quit jon-kha (anthony.freenode.net irc.freenode.net)
15:26:20 Quit w1ll14m|_ (anthony.freenode.net irc.freenode.net)
15:26:20 Quit elcan (anthony.freenode.net irc.freenode.net)
15:26:20 Quit AlexP (anthony.freenode.net irc.freenode.net)
15:26:20 Quit aevin (anthony.freenode.net irc.freenode.net)
15:26:20 Quit tha (anthony.freenode.net irc.freenode.net)
15:26:20 Quit blithe (anthony.freenode.net irc.freenode.net)
15:26:20 Quit scorche|sh (anthony.freenode.net irc.freenode.net)
15:26:20 Quit ChanServ (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Rob2222 (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Zagor (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Lss (anthony.freenode.net irc.freenode.net)
15:26:20 Quit HBK (anthony.freenode.net irc.freenode.net)
15:26:20 Quit linuxguy3 (anthony.freenode.net irc.freenode.net)
15:26:20 Quit jvd (anthony.freenode.net irc.freenode.net)
15:26:20 Quit n17ikh (anthony.freenode.net irc.freenode.net)
15:26:20 Quit jfc (anthony.freenode.net irc.freenode.net)
15:26:20 Quit kadoban (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Dhraakellian (anthony.freenode.net irc.freenode.net)
15:26:20 Quit pjm0616 (anthony.freenode.net irc.freenode.net)
15:26:20 Quit jordan` (anthony.freenode.net irc.freenode.net)
15:26:20 Quit tarbo (anthony.freenode.net irc.freenode.net)
15:26:20 Quit killan_ (anthony.freenode.net irc.freenode.net)
15:26:20 Quit gtkspert_ (anthony.freenode.net irc.freenode.net)
15:26:20 Quit crwl (anthony.freenode.net irc.freenode.net)
15:26:20 Quit Overand (anthony.freenode.net irc.freenode.net)
15:26:20 Quit courtc (anthony.freenode.net irc.freenode.net)
15:27:07NHealanthony.freenode.net irc.freenode.net
15:27:07NJoinChanServ [0] (ChanServ@services.)
15:27:07NJoinr00s [0] (n=ru@zentrale.profitables.biz)
15:27:07NJoinrobin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
15:27:07NJoinLinusN [0] (n=linus@rockbox/developer/LinusN)
15:27:07NJoinesperegu [0] (n=quassel@s559081d2.adsl.wanadoo.nl)
15:27:07NJoinmerbanan [0] (n=banan@c-83-233-172-245.cust.bredband2.com)
15:27:07NJoindaurn [0] (i=daurnima@freenode/staff/daurnimator)
15:27:07NJoindfkt [0] (i=dfkt@unaffiliated/dfkt)
15:27:07NJoinchb [0] (n=1000@unixboard/mod/chb)
15:27:07NJoinfunman [0] (n=fun@rockbox/developer/funman)
15:27:07NJoinuflops [0] (n=yogurt@90-231-195-226-no112.tbcn.telia.com)
15:27:07NJoineinhirn [0] (n=Miranda@bsod.rz.tu-clausthal.de)
15:27:07 Join rvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
15:27:07NJoinTheSeven [0] (n=theseven@dslb-084-056-142-155.pools.arcor-ip.net)
15:27:07NJoinmatsl [0] (n=matsl@91.191.140.126)
15:27:07NJoinDerPapst [0] (n=DerPapst@p4FE8FF56.dip.t-dialin.net)
15:27:07NJoinJdGordon [0] (n=jonno@rockbox/developer/JdGordon)
15:27:07NJoinTopy44 [0] (n=Topy44@g228147214.adsl.alicedsl.de)
15:27:07NJoinRob2222 [0] (n=Miranda@p4FDCCD88.dip.t-dialin.net)
15:27:07NJoinZagor [242] (n=bjorn@rockbox/developer/Zagor)
15:27:07NJoinshaggy-h [0] (n=kiwi@host-87-74-127-193.dslgb.com)
15:27:07NJoinflydutch [0] (n=flydutch@host77-167-dynamic.15-87-r.retail.telecomitalia.it)
15:27:07NJoinstoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
15:27:07NJoinmc2739 [0] (n=mc2739@cpe-67-10-237-133.satx.res.rr.com)
15:27:07NJoinLss [0] (n=Lss@cm46.delta91.maxonline.com.sg)
15:27:07NJoinkillan_ [0] (n=nnscript@c-0efa70d5.06-397-67626721.cust.bredbandsbolaget.se)
15:27:07NJoinRondom [0] (n=Rondom@dslb-084-057-154-162.pools.arcor-ip.net)
15:27:07NJoinpixelma [0] (i=quassel@rockbox/staff/pixelma)
15:27:07NJoinamiconn [0] (i=quassel@rockbox/developer/amiconn)
15:27:07NJoinmaraz_ [0] (i=maraz@xob.kapsi.fi)
15:27:07NJoinGodEater [0] (n=bibble@rockbox/staff/GodEater)
15:27:07NJoinsbhsu [0] (n=a6530466@Zion.dorm.au.edu.tw)
15:27:07NJoinBagder [241] (n=daniel@rockbox/developer/bagder)
15:27:07NJoinsjh [0] (n=sjh@svana.org)
15:27:07NJoinFlynDice [0] (n=FlynDice@c-24-19-225-90.hsd1.wa.comcast.net)
15:27:07NJoinparafin [0] (i=parafin@paraf.in)
15:27:07NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
15:27:07NJoinpolobricolo [0] (n=paul@AGrenoble-257-1-118-251.w90-27.abo.wanadoo.fr)
15:27:07NJointomers [0] (n=IceChat7@bzq-84-109-85-100.red.bezeqint.net)
15:27:07NJoinantil33t [0] (n=Mudkips@119.224.12.185)
15:27:07NJoinAndyI [0] (n=pasha_in@212.14.205.32)
15:27:07NJoinavacore [0] (i=nobody@1008ds1-rdo.0.fullrate.dk)
15:27:07NJoinDaveDavenport [0] (n=qball@ipd50a4125.speed.planet.nl)
15:27:07NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
15:27:07NJoinaidy [0] (n=aidy@216.94.9.114)
15:27:07NJoingtkspert_ [0] (n=gtkspert@124-169-3-222.dyn.iinet.net.au)
15:27:07NJoinHBK [0] (n=hbk@rrcs-97-77-51-170.sw.biz.rr.com)
15:27:07NJoinhd [0] (i=jd@Wikipedia/HellDragon)
15:27:07NJoinSajber^ [0] (n=Sajber@h-142-237.A213.priv.bahnhof.se)
15:27:07NJoinbubsy [0] (n=bubsy@94.139.72.137)
15:27:07NJoinlinuxguy3 [0] (n=timj@adsl-68-253-209-176.dsl.emhril.ameritech.net)
15:27:07Mode"#rockbox +o ChanServ " by irc.freenode.net
15:27:07NJoinShaid [0] (i=adam@174.143.241.229)
15:27:07NJoinkrazykit [0] (n=kkit@c-24-218-166-241.hsd1.ma.comcast.net)
15:27:07NJoinTorne [0] (i=torne@81.171.46.118)
15:27:07NJoinjvd [0] (n=syscrash@poipu/developer/syscrash)
15:27:07NJoinBeta2K [0] (n=beta@d24-36-68-97.home1.cgocable.net)
15:27:07NJoingoffa [0] (n=goffa@70.33.8.114)
15:27:07NJoinniekie [0] (i=quasselc@dreamworld.bergnetworks.com)
15:27:07NJoinDirtySkeMe [0] (n=Adium@nor75-19-82-244-50-203.fbx.proxad.net)
15:27:07NJoinn17ikh [0] (n=n17ikh@host-69-59-126-212.nctv.com)
15:27:07NJoinjfc [0] (n=john@dpc6682208002.direcpc.com)
15:27:07NJoinalexbobp [0] (n=alex@66.112.249.119)
15:27:07NJoinjanne_oksanen [0] (n=janne@130.234.191.99)
15:27:07NJoinjordan` [0] (i=gromit@78.235.252.137)
15:27:07NJoinkadoban [0] (n=mud@cpe-24-93-17-195.rochester.res.rr.com)
15:27:07NJoinDhraakellian [0] (n=ntryon@cpe-72-226-197-191.rochester.res.rr.com)
15:27:07NJoinHadaka [0] (n=naked@naked.iki.fi)
15:27:07NJoinAB3JU [0] (n=dz@alt.dissonance.nl)
15:27:07NJoinlyngaas [0] (n=staale@19.81-167-149.customer.lyse.net)
15:27:07NJoinDaCapn [0] (i=dacapn@using.your.wireless-inter.net)
15:27:07NJoinrwong [0] (n=ricky@www.roflwaffle.com)
15:27:07NJoinmeermanr [0] (n=meermanr@robmeerman.co.uk)
15:27:07NJoinJ-23 [0] (n=zelazko@unix.net.pl)
15:27:07NJoincg_ [0] (n=cromos@cable-kmi-fe71de00-186.dhcp.inet.fi)
15:27:07NJoinaevin [0] (i=eivindsy@unaffiliated/aevin)
15:27:07NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
15:27:07NJoinelcan [0] (i=user36@64.32.20.56)
15:27:07NJoinsinthetek [0] (n=sinthete@cpe-075-183-051-184.triad.res.rr.com)
15:27:07NJoinpreglow [0] (i=thomj@tvilling2.pvv.ntnu.no)
15:27:07NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
15:27:07NJoinCIA-43 [0] (n=CIA@208.69.182.149)
15:27:07NJoinzu [0] (n=zu@bucketheaded.eu)
15:27:07NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
15:27:07NJoinjon-kha [0] (i=jon-kha@83.150.91.127)
15:27:07NJoinblithe [0] (n=blithe@72.14.176.144)
15:27:07NJointha [0] (i=1038@130.83.160.206)
15:27:07NJointarbo [0] (n=me@unaffiliated/tarbo)
15:27:07NJoinpjm0616 [0] (n=user@61.250.113.98)
15:27:07NJoinw1ll14m|_ [0] (n=nobody@84.104.81.17)
15:27:07NJoinThomasAH [0] (n=thomas@aktaia.intevation.org)
15:27:07NJoinGalois [0] (i=djao@efnet.math.uwaterloo.ca)
15:27:07NJoinAlexP [0] (n=alex@rockbox/staff/AlexP)
15:27:07NJoinyosafbridge [0] (n=yosafbri@ludios.net)
15:27:07NJoinrphillips [0] (n=rphillip@66-90-184-91.dyn.grandenetworks.net)
15:27:07NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
15:27:07NJoinOverand [0] (i=overand@crappy.domain.name)
15:27:07NJoincrwl [0] (n=crwlll@a91-156-100-168.elisa-laajakaista.fi)
15:27:07NJoincourtc [0] (n=court@unaffiliated/courtc)
15:27:39NJoinps-auxw [0] (n=arneb@dyn37.ps-auxw.de)
15:27:39NJoinmarkun [50] (n=markun@rockbox/developer/markun)
15:28:30 Join chandoo [0] (n=chandoo@ool-4353b978.dyn.optonline.net)
15:31:15NJoinpcc1 [0] (n=peter@master.pcc.me.uk)
15:32:08likemindead0_o
15:35:47 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
15:36:01CtcpVersion from freenode-connect!freenode@freenode/bot/connect
15:38:21 Quit matsl (Read error: 145 (Connection timed out))
15:40:09 Join rasher [0] (n=rasher@0x5550f5a3.adsl.cybercity.dk)
15:41:19 Quit fxb (Read error: 54 (Connection reset by peer))
15:41:37 Quit chb (Read error: 110 (Connection timed out))
15:41:57 Quit z35 (Read error: 104 (Connection reset by peer))
15:46:48 Quit sjh (Read error: 110 (Connection timed out))
15:47:49 Join fxb__ [0] (n=felixbru@h1252615.stratoserver.net)
15:47:57 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net)
15:49:14 Join jgarvey [0] (n=jgarvey@cpe-098-026-065-013.nc.res.rr.com)
15:49:59 Join Ypsy [0] (n=ypsy@geekpadawan.de)
15:50:04YpsyAloha
15:50:52YpsyI've got a few questions to the Fuze (v1) version. Is it still possible to boot the usual firmware afterwards? and what does "USB -> no" mean?
15:51:26funmanyes dual boot is possible, and usb is not handled by rockbox (the fuze reboots to OF when usb is plugged)
15:52:08notlistening:P funman types faster than me ;)
15:52:08Ypsyah okay so for filling it with music and charging i'll just boot the normal firmware
15:52:41 Join AndyIL [0] (n=pasha_in@212.14.205.32)
15:53:41Ypsywould it be a good idea to format and reset the player before starting?
15:58:40mc2739Ypsy: you do nit need to reboot to the OF for charging. It is not necessary to format your player before install, but you can if you prefer
15:58:52mc2739nit = not
15:58:58Ypsyokay
15:59:17Ypsythen i'll now give it a try :)
16:00
16:01:10mc2739Ypsy: In Rockbox, you can charge while using your player by pressing a button while inserting your USB cable. On the Fuze, it is the select (center) button
16:01:32Ypsyah nice :)
16:01:54Ypsythis isnt working with the of, is it? :P
16:02:36mc2739no, this is a feature of Rockbox
16:05:56 Quit esperegu (Read error: 104 (Connection reset by peer))
16:06:18 Join esperegu [0] (n=quassel@s559081d2.adsl.wanadoo.nl)
16:06:22 Quit AndyI (Read error: 110 (Connection timed out))
16:08:56 Join evilnick [0] (i=0c140464@gateway/web/freenode/x-mohsbautvdlytand)
16:11:13 Join Omlet [0] (n=omlet05@91.182.39.148)
16:11:23 Join Omlet05 [0] (n=omlet05@91.182.39.148)
16:36:42 Join toffe82 [0] (n=chatzill@12.169.218.14)
16:37:40 Join daurn| [0] (n=daurnima@freenode/staff/daurnimator)
16:43:47Ypsyisnt it possible to use the quickstart options for the fuze?
16:44:00Ypsy(in the rb utility)
16:52:38 Quit daurn (Read error: 110 (Connection timed out))
16:58:02 Join Strife89 [0] (n=michael@168.16.239.253)
17:00
17:01:04 Quit daurn| (Read error: 110 (Connection timed out))
17:01:49 Join daurn| [0] (n=daurnima@freenode/staff/daurnimator)
17:05:59 Quit mt (Read error: 101 (Network is unreachable))
17:06:00***Saving seen data "./dancer.seen"
17:14:30 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
17:14:59*TheSeven potentially found that memory corruption bug on nano2g
17:15:29TheSevenrockbox has different memset() arguments than apple
17:21:00funmanTheSeven: memset() has the same prototype on rockbox than on C89?
17:21:47TheSevenrockbox: memset(buffer, value, size);
17:21:58TheSevenapple (and thus me): memset(buffer, size, value);
17:22:53gevaertswhat? Apple has size,value?
17:23:17Zagorposix is overrated...
17:23:25gevaertsthat's not posix, that's C
17:23:36Zagorindeed :)
17:23:39gevaertsvoid *memset(void *s, int c, size_t n);
17:23:56*gevaerts copied that straight from the C99 spec
17:23:56funmangevaerts: nano2g firmware doesn't claim any C compliance
17:24:50*likemindead is away: FAILURE −− Sometimes your best just isn't good enough.
17:25:13funmanlikemindead: please avoid those messages here
17:25:30TheSevenwell, it at least confused me quite a bit as i was used to seeing it the wrong way round all the way disassembling the OFW
17:26:02Tornelots of embedded systems use it the wrong way around, yes
17:26:11Tornebecause it's easier to do bzero() then
17:26:45gevaertsthat's not a good reason
17:27:03Tornei didn't say it was a *good* reason
17:27:35 Part LinusN
17:27:38*gevaerts suspect that people just did it wrong and then tried to rationalise it
17:29:54*Torne has seen compilers which generate calls to a support function which is usually called something else but is memset with the 2nd/3rd args reversed even. :(
17:33:03TheSevenTorne: for zeroing out structs and such stuff?
17:33:16*TheSeven really needs to look at a disassembly of copying a struct
17:33:43 Quit Zagor ("Don't panic")
17:33:52*gevaerts doesn't mind if people have a function that uses this order, but if they call it memset() they should be fired
17:35:25 Quit Sajber^ ("Leaving.")
17:36:27TorneTheSeven: yeah, that kind of thing
17:36:51Tornegevaerts: unfortunately when i'm disassembling something and i find a functio nthat sets a range of memory to a particular value i name it memset even if the args are not the same ;)
17:37:11gevaertsTorne: you should name it ()tesmem
17:37:22Tornebut it's not backwards either! :)
17:37:34TheSeventhe arg's aren't backwards either, too
17:37:48TheSevenTorne: honestly, you should name it broken_memset or weird_memset, to not confuse people reading that disassembly later
17:38:14TheSevenmetesm() :-D
17:40:12 Join Sajber^ [0] (n=Sajber@85.24.142.150)
17:40:30Torneso, er, did we come to a conclusion about the usb charging stuff?
17:40:59TorneI'm gonna see about supporting the "proper" behaviour on USB hosts (controlled by a setting)
17:41:16Torneis it worth me figuring out how to detect the ac adapter, though, if people are going to want to ignore that anyway?
17:41:19 Quit Omlet (Read error: 60 (Operation timed out))
17:42:04 Join _lifeless [0] (n=lifeless@89.20.117.209)
17:42:43 Quit JdGordon ("Leaving.")
17:43:53 Join saratoga [0] (i=9803c6dd@gateway/web/freenode/x-lklytxvqbpsofbwg)
17:44:28 Quit Omlet05 (Read error: 145 (Connection timed out))
17:45:15TheSevenhost / non-host is the more important decision than if an ac adaptor complies to the one or another standard
17:45:39TheSevenbtw, what's the recommended way to flush the data cache in a rockbox driver?
17:46:51saratogayou mean to write the cache_flush functions
17:46:55TorneTheSeven: well I was gonna just do it for hosts for now
17:47:20TorneTheSeven: on the basis that that patch should be fine/safe/etc for everyone and it will let the ipods charge properly for most people
17:47:39TheSevensaratoga: where do these functions belong?
17:47:51TheSeventhat mmu-arm.c isn't really applicable to the s5l8700
17:47:54Tornei guess the most complete way would be to also detect the ac adapter via any method that works..
17:47:56saratogaTheSeven: they're usally in the target tree for each CPU family
17:48:05Torneand then *still* have a setting to allow charging anyway
17:48:09saratogabut since they're generic to most arm chips of a given version, you may not need to write them
17:48:33 Join dash32 [0] (n=dash32@p54AB7BF8.dip.t-dialin.net)
17:48:41gevaertsTorne: I agree. Make sure that the ipod charger detection can be added later :)
17:49:08TheSevensaratoga: so you would say i should place them in target/arm/s5l8700/pmu-s5l8700.h?
17:49:11saratogaI think all ARM9 chips for instance have the same cache flush commands
17:49:20saratogayes that would be fine I think
17:49:25saratogaif you're not sure look at how AMS does it
17:49:33Torneyes, ARM cache stuff is generic
17:49:40saratogaits just a CP15 command right?
17:49:52Tornevery generic, in fact; ARM7/9/11/A8/etc all use the same CP15 commands modulo minor intresting things
17:50:11Torne(like XScale being weird, and old chips not having harvard caches)
17:50:31TheSevenyep, but it still doesn't work, as that mmu-arm.c thing references things, that simply don't exist for ARM9
17:50:41*likemindead is back (gone 00:25:50)
17:50:48TheSevenshould I ifdef that out?
17:51:01TheSevenlikemindead: please read http://www.rockbox.org/wiki/IrcGuidelines
17:51:23TorneTheSeven: like what?
17:51:38saratogaARM7 isn't harvard fwiw
17:51:51likemindeadGuidelines noted. I will use them as a guide.
17:52:19Tornesaratoga: up to the person synthing it, no?
17:52:35saratogaAFAIK they're all Von Neuman
17:52:45Tornebut probably not relevant for us since the only arm7 targets we have are arm7tdmi, no?
17:52:48Tornewhich has no cache
17:52:50TheSevenlikemindead: then please configure your IRC client accordingly, to not post such things
17:52:53Torneeven on the chips where it does :)
17:53:03Torne(portalplayer)
17:53:43saratogaI think all ARM7 have a single wide load pathway so i don't think you could make them harvard even if you wanted to, except maybe in thumb mode
17:54:06saratogathough i guess no one is prevented from redesigning the arm7 front end to be wider, though i don't know if its still arm7 if you redesign it
17:54:24TheSeventhe issue was more-advanced mmu stuff in there requiring some reg defines, which just don't exist on the s5l8700, as it doesn't support segmentation for example
17:54:44 Quit petur ("work->home")
17:55:44Tornethe TTB stuff is for MPU arches
17:55:57likemindeadBy definition, following a guideline is never mandatory (protocol would be a better term for a mandatory procedure).
17:56:00Torneif you have an MMU you want the code that imx31l uses, modulo one or two instructions..
17:56:07likemindeadhttp://en.wikipedia.org/wiki/Guidelines
17:56:09likemindead;-)
17:56:36saratogaok how about you just stop being annoying and we agree not to ban you for it
17:56:45TheSevenTorne: the non-imx code, minus the ttb stuff, matches better
17:57:08Torneis it an mmu or mpu?
17:57:11gevaertslikemindead: not banning you is also not mandatory
17:57:26likemindead:::rolls eyes:::
17:57:28 Part likemindead ("FAILURE -- Sometimes your best just isn't good enough.")
17:57:45TheSevenhe doesn't know /me?
17:58:25Tornethey're pretty similar anyway, no?
17:58:53 Quit shaggy-h ()
17:58:54Torneoh, no, i'm looking at the wrong bit..
17:59:21Tornesorry, yeha, the imx one uses the byrange cache funcs which are only on armv6
17:59:51 Quit stoffel (Remote closed the connection)
18:00
18:02:08 Join Grahack [0] (n=chri@ip-222.net-82-216-222.rev.numericable.fr)
18:08:25 Quit Strife89 ("Lunchtime.")
18:08:37 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
18:12:20 Quit dfkt (Read error: 110 (Connection timed out))
18:19:50TheSevenTorne: so what to do about it? ifdef that TTB stuff somehow?
18:20:14*TheSeven is angry that his current rockbox build hangs while booting up
18:21:36 Quit stoffel (Remote closed the connection)
18:24:08 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
18:24:20 Join evilnick_ [0] (n=evilnick@ool-4571af51.dyn.optonline.net)
18:24:29 Join howell_ [0] (n=howell@rrcs-97-77-48-15.sw.biz.rr.com)
18:25:09 Quit scorche (Nick collision from services.)
18:25:21 Join Unhelpful_ [0] (n=quassel@pool-71-173-205-32.hrbgpa.fios.verizon.net)
18:25:37 Join bzed_ [0] (n=bzed@devel.recluse.de)
18:25:55 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
18:26:26 Quit howell7 (anthony.freenode.net irc.freenode.net)
18:26:26NSplitanthony.freenode.net irc.freenode.net
18:26:26 Quit evilnick_home (anthony.freenode.net irc.freenode.net)
18:26:26 Quit advcomp2019 (anthony.freenode.net irc.freenode.net)
18:26:26 Quit FOAD (anthony.freenode.net irc.freenode.net)
18:26:26 Quit JackWinter (anthony.freenode.net irc.freenode.net)
18:26:26 Quit shadearg (anthony.freenode.net irc.freenode.net)
18:26:26 Quit Unhelpful (anthony.freenode.net irc.freenode.net)
18:26:26 Quit Res1 (anthony.freenode.net irc.freenode.net)
18:26:26 Quit crashd (anthony.freenode.net irc.freenode.net)
18:26:26 Quit Kohlrabi (anthony.freenode.net irc.freenode.net)
18:26:26 Quit bzed (anthony.freenode.net irc.freenode.net)
18:26:41 Nick bzed_ is now known as bzed (n=bzed@devel.recluse.de)
18:27:04NHealanthony.freenode.net irc.freenode.net
18:27:04NJoinJackWinter [0] (n=jack@vodsl-10103.vo.lu)
18:27:51 Join bertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl)
18:28:40 Join crashd [0] (i=foobar@lostnode.org)
18:28:50NJoinFOAD [0] (n=dok@dinah.blub.net)
18:28:59NJoinKohlrabi [0] (n=Kohlrabi@frustrum.nosebud.de)
18:29:10 Quit LambdaCalculus37 ("Fwump")
18:32:17*TheSeven managed to port the new FTL to rockbox now
18:32:34TheSevenlet's stress test it a little to see if those bugs have gone away
18:37:56 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:39:20 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
18:40:29 Quit esperegu (Read error: 113 (No route to host))
18:43:12TheSevenno, that didn't fix the data corruption issue
18:43:23 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
18:44:14NJoinshadearg [0] (i=arg@ipv4.panoptix.net)
18:47:09 Quit JackWinter (Remote closed the connection)
18:48:01 Nick Ypsy is now known as YPSY (n=ypsy@geekpadawan.de)
18:48:54 Join JackWinter [0] (n=jack@vodsl-10103.vo.lu)
18:53:14Mode"#rockbox +o funman " by ChanServ (ChanServ@services.)
18:53:51 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
19:00
19:04:12 Join gb_master [0] (i=82c03212@gateway/web/freenode/x-bgiqlncvuyjdsjmj)
19:06:02***Saving seen data "./dancer.seen"
19:08:34 Quit gb_master (Client Quit)
19:09:10 Quit funman ("free(random());")
19:12:13TheSevendoes anybody know how the rockbox FAT driver handles blocksizes != 512?
19:18:23 Join mcuelenaere [0] (n=mcuelena@78.22.180.217)
19:20:41 Quit JackWinter (Read error: 104 (Connection reset by peer))
19:21:42TheSevenis there *anybody* here who has a clue about the FAT driver?
19:22:07gevaertsprobably
19:24:19gevaertsTheSeven: my guess is that amiconn is probably the one to ask
19:24:30*gevaerts bases this on r11659
19:25:17 Join JackWinter [0] (n=jack@vodsl-10103.vo.lu)
19:38:35 Quit ender` (" If Klingons had invented Usenet, killfiles really would...")
19:47:48 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
19:49:06saratogaanyone want to speculate why my patch for doing logf directly to disk works fine on the sim but only creates a blank file on AMS?
19:49:07saratogahttp://pastebin.com/m474692ce
19:49:16saratogai'm a little confused
19:55:56 Join Omlet [0] (n=omlet05@204.104-240-81.adsl-dyn.isp.belgacom.be)
19:55:58 Join Omlet05 [0] (n=omlet05@204.104-240-81.adsl-dyn.isp.belgacom.be)
19:57:30 Quit einhirn ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
19:57:51 Join Thundercloud [0] (i=thunderc@persistence.flat.devzero.co.uk)
19:58:56 Quit stoffel (Remote closed the connection)
20:00
20:00:25domonokysaratoga: did you check if the file open/close work/ are called correctly on target ?
20:01:16saratogano but i assumed they did
20:01:21saratogaperhaps I should try a more mature target
20:01:57domonokysaratoga: and also the check at line 181 in this patch is probably not what you want, you are comparing if the two char pointers are not the same. not their contents (but that shouldnt be the problem)
20:02:46saratogahmm good catch
20:03:01saratogai blame uchida for that one
20:03:29 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net)
20:05:38 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
20:08:19domonokysaratoga: where do you call logf_file_close() on sims ? i only see it in the target part of clean_shutdown(). And it may be called too late on target ?
20:09:22*TheSeven finally fixed that "NAND data corruption" issue
20:09:41TheSevenin fact it was a misunderstanding on how rockbox storage works on linuxstb's side :-)
20:12:49saratogadomonoky: it probably doesn't get called on the sim
20:13:03saratogai don't think thats a problem, you just don't get the end of log message printed
20:13:58saratogaalso wow theres a whole mess of wma optimizations posted to ffmpeg
20:14:10domonokynot closing a file, might corrupt it so you dont see any content i think.
20:18:09saratogahow would that happen?
20:22:25 Quit flydutch ("/* empty */")
20:22:28 Quit Omlet05 ("( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )")
20:22:28 Quit Omlet ("( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )")
20:34:04 Quit stoffel (Remote closed the connection)
20:38:45 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
20:40:07 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
20:46:51 Quit stoffel (Remote closed the connection)
20:48:18 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
20:49:15 Join Omlet [0] (n=omlet05@204.104-240-81.adsl-dyn.isp.belgacom.be)
20:50:09 Join Omlet05 [0] (n=omlet05@204.104-240-81.adsl-dyn.isp.belgacom.be)
20:55:57 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
20:57:25saratogawhat happens if i call strcmp on an empty char[] buffer?
20:58:58saratogai guess it won't be equal pretty quickly so it just terminates
20:59:35gevaertsif you're lucky
21:00
21:04:39 Quit n17ikh ()
21:06:01saratogahmm is "/.rockbox/logf.log" not a valid rockbox path?
21:06:05***Saving seen data "./dancer.seen"
21:07:48 Join matsl [0] (n=matsl@1-1-4-2a.mal.sth.bostream.se)
21:10:55TheSevendoes anybody know how those ipods with 1024 bytes per sector are formatted?
21:11:29TheSevenare they simulating 512 byte sectors and have a 512-based FS on them, or are they exposing the 1024 byte sectors?
21:11:40TheSeven(and thus have a 1024-byte-FS on them)
21:11:53gevaertsthey actually have a 2048 byte FS on them
21:12:27TheSeven2048 bytes sector size or cluster size?
21:12:41gevaertssector
21:12:49TheSevenhow does this work then? fat.c looks like being hardwired to 512 to me
21:13:02gevaertsthey actually expose a 2048 byte sector size over USB
21:13:04TheSevenit includes fat.h and that defines SECTOR_SIZE as 512
21:13:29gevaertsindeed it does. That's there to confuse people!
21:13:43*gevaerts was confused by it when he did the MSC support
21:13:59TheSevenhow does it work then? some include path trick that makes it use another file?
21:14:40*gevaerts will now say some things that may well be totally wrong
21:16:10gevaertsThe storage drivers all expose 512 byte sectors, hence SECTOR_SIZE is 512. The ipod ata driver can recognise this 1024 byte sector disk, and on seeing it, it makes sure to do 1024-byte I/O (by e.g. reading 1024 bytes, overwriting 512 of them in RAM and writing the lot back when asked to write 512 bytes)
21:16:34gevaertsThe fat driver does some additional magic that I'm not familiar with...
21:17:10gevaertsIt's all caused by this strange ipod not being there when the drivers were first written
21:17:23TheSeveni just had a look at the fat driver and it looks to me like it can't ever work with SECTOR_SIZE being something different than assumed by the FS, as it directly passes the calculated sector numbers down to storage, without doing any magic on them
21:17:55 Join n17ikh [0] (n=n17ikh@host-69-59-126-212.nctv.com)
21:17:58TheSevenif I however define SECTOR_SIZE as 2048 and expose 2048-byte-sectors through storage, everything works just fine
21:17:58 Quit Omlet ("( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )")
21:17:59 Quit Omlet05 ("( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )")
21:19:31 Join stripwax [0] (n=Miranda@87-194-34-169.bethere.co.uk)
21:21:15*gevaerts doesn't know more about this
21:21:41saratogadomonoky: so I tried hardcoding a string write immediately after opening the file
21:21:50saratogaand the file gets created but the write never happens
21:22:59domonokysaratoga: strange.perhaps try with open-write-close to see if that works.
21:26:16saratogatrying that now
21:28:02 Quit robin0800 (Remote closed the connection)
21:28:58saratogahuh that works
21:29:10saratogaso it must be the close
21:31:41 Join petur [0] (n=peter@rockbox/developer/petur)
21:34:31 Quit thegeek (Read error: 104 (Connection reset by peer))
21:34:31 Join thegeek [0] (n=nnscript@s168c.studby.ntnu.no)
21:34:31 Join Zarggg_ [0] (n=zarggg@65-78-69-194.c3-0.eas-ubr6.atw-eas.pa.cable.rcn.com)
21:34:31 Quit Zarggg (Read error: 104 (Connection reset by peer))
21:40:43 Join liar|netbook [0] (n=liar@83.175.83.185)
21:50:12 Quit stoffel (Remote closed the connection)
21:50:45 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
21:51:10 Quit stripwax ("http://miranda-im.org")
21:57:40 Quit LambdaCalculus37 ("Fwump")
22:00
22:00:17 Quit Thundercloud (Remote closed the connection)
22:00:37 Join XXXX [0] (n=XXXXXXXX@ama-wldhcp.69614189.amaonline.com)
22:00:41 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
22:02:02 Quit stoffel (Remote closed the connection)
22:02:44 Nick YPSY is now known as Ypsy (n=ypsy@geekpadawan.de)
22:06:32TheSeventhe rockbox FAT driver is buggy
22:06:46TheSevenit panics, while trying to fiddle with some nonexistant files.
22:07:37 Join stoffel [0] (n=quassel@p57B4EA99.dip.t-dialin.net)
22:10:52TheSeveni think it's trying to create some file in .rockbox, but it doesn't work somehow
22:11:03 Join Paprica [0] (n=a@93-172-61-190.bb.netvision.net.il)
22:11:07 Quit stoffel (Remote closed the connection)
22:11:17 Join kugel [0] (n=kugel@rockbox/developer/kugel)
22:11:53Papricahi, someone knows how to play spider splitare? I need tasters for my plugin XD
22:12:19 Join ender` [0] (i=krneki@foo.eternallybored.org)
22:14:27 Join Zagor [242] (n=bjst@46.35.227.87.static.tab.siw.siwnet.net)
22:19:26 Quit merbanan (Read error: 110 (Connection timed out))
22:25:18 Quit XXXX (Read error: 110 (Connection timed out))
22:25:21TheSevenamiconn, gevaerts, whoever: is there any sensible reason for keeping SECTOR_SIZE=512? it seems to work just fine with 2048, and everything else leads to a huge overhead in both performance and flash wear, and doesn't even seem to work properly
22:25:38 Join XXXX [0] (n=XXXXXXXX@ama-wldhcp.69614189.amaonline.com)
22:27:52 Quit JackWinter (Remote closed the connection)
22:28:17gevaertsTheSeven: I suspect that it has to remain 512 on any target that can ever meet 512 byte sector filesystems. The nano presumably isn't one of those though
22:28:32TheSevenyep, that seems to be true
22:28:38TheSeveni'll go for a
22:28:52TheSeven#ifndef SECTOR_SIZE
22:28:54TheSeven#define SECTOR_SIZE
22:28:56TheSeven#endif
22:28:57 Join JackWinter [0] (n=jack@vodsl-10103.vo.lu)
22:29:07 Part Grahack
22:29:10TheSevenconstruct now, and define SECTOR_SIZE = 2048 in config-ipodnano2g.h
22:29:15gevaertsof course there could be some hidden bugs if you change it, but that shouldn't be a reason not to change things
22:29:35notlisteningdomonoky, http://www.rockbox.org/tracker/task/10312 enjoy ;) Just writing the development notes on my wiki see the post.
22:29:40TheSevenyep, i'm more or less trying to bring those up - so far caught only one hardwired 512
22:30:18gevaertsI know there's one in usb_storage.c, where I defined my own SECTOR_SIZE...
22:30:27*gevaerts can't remember why he ever did that
22:30:46TheSeventhere's another 512 hardcoded in there btw, that isn't even using your SECTOR_SIZE
22:31:01gevaertsprobably because I didn't want to have anything to do with FAT, but that can also be solved by moving that to the config system
22:31:03TheSevenjust spotted that accidentally, the one bugging me was in disk.c
22:31:20gevaertsoh indeed. Bad me!
22:32:29 Join Grahack [0] (n=chri@ip-222.net-82-216-222.rev.numericable.fr)
22:33:05 Quit Grahack (Client Quit)
22:33:29*TheSeven wonders what led() is doing in nano2, it's used in ata-nand-s5l8700.c
22:33:55CIA-43New commit by 03gevaerts (r22869): get rid of one hardcoded 512. SECTOR_SIZE is still there. It should probably move up (to storage.h?) so fat and usb storage can share it
22:35:08TheSevengah, another conflict for my patch ;-)
22:35:25TheSevenah, wait, it was the other 512, so it shouldn't hurt
22:36:29 Join Grahack [0] (n=chri@ip-222.net-82-216-222.rev.numericable.fr)
22:36:50 Quit Grahack (Client Quit)
22:37:11TheSeven"ATA error: 1\n\nPress ON to debug"
22:37:15TheSevenwhat the hell does this mean?
22:37:55gevaertsit means you met old code :)
22:38:05TheSevenany idea where it is?
22:38:30Zagorthat's in main, isn't it?
22:39:12TheSevenyep, grep just caught it
22:40:29domonokynotlistening: great. so i know what todo this weekend :-)
22:41:27notlisteningdomonoky, if i can help anyways let me know
22:41:44notlisteningI am off to a ubuntu Jam :?
22:42:49bertrikoh the red fixed itself
22:43:06domonokynotlistening: will do. have fun at the jam :-)
22:43:35notlisteningthanks ;)
22:47:51amiconnTheSeven: The 80GB 5.5G has 2048 byte logical FAT sectors, 512 byte logical ATA sectors, and 1024 byte physical ATA sectors. Confused?
22:48:24 Join Strife89 [0] (n=michael@168.16.232.201)
22:48:30TheSevenonly by one point: what is logical != physical good for?
22:49:17gevaertsTheSeven: the disk is addressed in 512-byte sectors, but can't handle odd addresses
22:49:22 Quit Strife89 (Client Quit)
22:49:28gevaerts(more or less, I hope)
22:49:36amiconnAlso, the disk's firmware doesn't handle partial sector accesses, i.e. while you're always addressing it based on 512-byte sectors, you must start at even sector numbers (counting 0-based) and read even numbers, otherwise the disk sets IDNF
22:50:28TheSevenwhy don't you just emulate that as 1024-byte sectors, even though you might need to multiply the internal sector number by 2?
22:50:56amiconnThat would be difficult
22:50:56TheSevenwhat is the downside of keeping it simple that way?
22:51:13kugelbertrik: I tried your dbop patch again. I'm totally out of ideas why hold doest work (I tried adding a delay and changing DBOP_CTRL bits a bit)
22:51:14TorneTheSeven: The ATA spec doesn't allow for it
22:51:27TorneTheSeven: it was easier to keep soem modicum of compatibility this way
22:51:32amiconnIt would mean that you *always* have to read and write 1024 byte sectors, but the buffers in the FAT driver only hold 512 byte sectors
22:52:01gevaertsTheSeven: not all ipod video disks behave like that
22:52:11amiconnYeah, that's another point - ATA always addresses in multiples of 512 bytes, regardless of physical sector size
22:52:13TheSevenyep, that's the point: *why* do they only hold 512 and not 2048 bytes, as the FS is based on 2048 byte sectors?
22:52:37TheSevenwhat you're doing at ATA driver level has nothing to do with that
22:52:41amiconnThe buffers are often on stack, and stack is limited
22:53:11amiconnAlso, this driver is shared across all targets, and low memory usage is highly desired
22:53:36TheSeveni don't know of low-memory targets that don't have 512-byte sectors
22:53:47amiconnSo far the MK8011GAH is the *only* disk which is unable to handle partial accesses we know about
22:54:03TheSevenand almost every NAND chip
22:54:27TorneTheSeven: the natural sector size for a NAND chip is the eraseblock size, though, not the page size
22:54:38 Join FOAD_ [0] (n=dok@dinah.blub.net)
22:54:40Torneif only FSes didn't choke on that :)
22:54:48Torneit would make ftl much easier
22:55:01TheSevenwell, using the write block (i.e. page) size is a sensible way to get around this
22:55:36TheSevenand a FS with 1MB sectors would be neither space nor wear efficient
22:55:36amiconnThere's a newer, large capacity 1.8" hdd (forgot the type) which even has 4096 byte physical sectors. That one is working fine without any physical sector magic (in fact rockbox needs to be compiled without it in order too work)
22:55:36TheSevenFTLs do a much better job there
22:55:37TheSevenand obviously adjusting this per-target isn't an issue, so for now I'll just go for SECTOR_SIZE=2048 on nano2g
22:56:02*amiconn thinks the ata driver should actually probe for partial sector support, and only apply its own magic when the disk complains
22:56:09Torneamiconn: surely it should in theory work better if we *did* treat it as 4096 sectors
22:56:13Torneer, bytes
22:56:20amiconnIt doesn't
22:56:23 Quit bmbl ("Bye!")
22:56:26Tornethat's.. weird.
22:56:30amiconnIt's in fact worse
22:56:49bertrikkugel, did you try "pre-charging" it both as 1 and 0?
22:56:57kugelyep
22:57:10bertrikalso I was thinking that maybe this pins is connected to another pin
22:57:25bertrikI think I should just have a closer look at how the OF does it
22:57:41amiconnIf the disk does it, the disk *knows* that more partial sectors could be transferred from the host, so it delays writing a bit. However, if rockbox has to do it, the ata driver doesn't know this, or at tleast it can't be guaranteed
22:57:43kugelour SVN code is based on the OF code
22:58:20amiconnHence the driver has to insert each partial sector in a full sector, writing each full sector several times (not reading though, since that is cached)
22:58:27TheSevenok, there are now 2 storage bugs left for nano2g: Something is still wrong with writes, and with dcache enabled something goes terribly wrong
22:58:41amiconnThis causes the disk head to constantly jump back& forth a bit - *extremely* slow
22:59:15Torneamiconn: ah, right. so it's worse because we don't cache writes, pretty much?
22:59:20Torneit would be better for an OS which did, though..
22:59:31kugelTheSeven: are you using dma of some sort?
22:59:45TheSeven yep, and in theory i clean the caches properly, but it just won't work
22:59:49amiconnThe only exception is when writing actual data clusters - those are transferred directly from memory, in full size (up to 128KB, the maximum possible ata transfer size w/o LBA48 support)
23:00
23:00:02kugelthe samsas had horrible problems with sd transfers when initially enabling dcache
23:00:36*pixelma wonders what is so hard about spelling out AMS sansa and without
23:01:15kugelcleaning/flushing wasn't enough. the dma buffer needs to be uncached if the address is not cache-aligned (I can't tell it in detail)
23:01:36kugelFlynDice and funman know more about the problems we had with dcache
23:02:31 Join Thundercloud [0] (i=thunderc@persistence.flat.devzero.co.uk)
23:02:35amiconnTorne: Caching writes is calling for real trouble
23:02:44Torneamiconn: yes, i'm not suggesting we *do* this
23:02:58Tornei am just checking i understood you
23:03:14Torneif we were a desktop OS that did cache writes it probably would be better to follow the physical sector size, but we don't, so it's not.
23:04:22 Quit dash32 ("Verlassend")
23:04:32TheSevenkugel: can you quickly name a target that has this fixed, so i can check how it was done??
23:04:49amiconnAlso, SECTOR_SIZE != 512 means opening a can of worms. I expect all sorts of bugs creeping up
23:05:04kugelall samsas (or AMS sansas...) see target/arm/as3525/sd_ata_as3525.c
23:05:20kugelmake sure to check the svn log
23:05:30TheSeven[23:03]<amiconn>Also, SECTOR_SIZE != 512 means opening a can of worms. I expect all sorts of bugs creeping up
23:05:38TheSevenone of the reasons why i'm doing this
23:05:52 Quit LambdaCalculus37 ("Fwump")
23:06:01kugelbut doing data cache in such an early state of a part seems like asking for trouble
23:06:08***Saving seen data "./dancer.seen"
23:06:43domonokycorrectly flushing caches with dma is really important :-)
23:06:47FlynDiceTheSeven: lsee also system as3525 and pcm-as3525
23:06:59FlynDicefor memory mappings
23:07:50FlynDiceand FS #10048 for the problem solving...
23:08:07 Quit XXXX (Remote closed the connection)
23:08:43 Join domonoky1 [0] (n=Domonoky@e179161023.adsl.alicedsl.de)
23:09:27 Quit evilnick ("Page closed")
23:10:35 Quit FOAD (Read error: 110 (Connection timed out))
23:10:35 Nick FOAD_ is now known as FOAD (n=dok@dinah.blub.net)
23:10:36 Join funman [0] (n=fun@rockbox/developer/funman)
23:12:05kugelFlynDice: can you tell if the scrollwheel is better or worse after my commit last night?
23:12:08TheSevenFlynDice: wow, nice trick!
23:13:01funmanFlynDice: why did i know this FS# by heart ? ;)
23:13:03FlynDicekugel: I can't tell better or worse but it's working fine right now...
23:14:56FlynDicefunman: Didn't it take as long to figure that out as for you to write the rest of the code?.... ;)
23:15:20 Quit bertrik ("De groeten")
23:15:51*kugel was hoping to see that FS# never ever again :)
23:16:23funmanFlynDice: i think even longer ;) Thanks again for helping on those major problems
23:16:39 Quit ender` (Read error: 104 (Connection reset by peer))
23:19:48 Quit petur ("Zzzzz")
23:19:54 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
23:21:07 Join ender` [0] (i=krneki@foo.eternallybored.org)
23:24:03 Quit jgarvey ("Leaving")
23:26:01 Quit domonoky (Read error: 110 (Connection timed out))
23:26:08 Join stripwax [0] (n=Miranda@87-194-34-169.bethere.co.uk)
23:31:12 Quit funman ("leaving")
23:31:35 Nick Ypsy is now known as YPSY (n=ypsy@geekpadawan.de)
23:39:58 Quit ender` (" It always takes longer than you expect, even when you take Hofstadter's Law into account. -- Hofstadter's Law")
23:46:44 Join Res1 [0] (n=Res@user-0c6s6ir.cable.mindspring.com)
23:49:02 Quit AlexP ("Please insert girder")
23:49:29 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
23:52:03 Join ender` [0] (i=krneki@foo.eternallybored.org)
23:52:50 Join robin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
23:55:31TheSevenYEAH! Read works now, even with dcache
23:56:56TheSevenbtw where is linuxstb hanging around again? haven't seen him for a week.
23:57:10AlexPHe'll be around when not busy
23:57:10*Bagder adds committer #81
23:57:13kugelTheSeven: did the as3525 code help?
23:57:16gevaerts\☺/
23:57:20kugel\o/
23:57:31TheSevenkugel: didn't use the actual code, but the idea did the trick :-)
23:57:33gevaertsdo we get real beer at #100?
23:57:48BagderI hope so!
23:57:54kugelTheSeven: the idea is often the most critical part of an implementation :)
23:58:23TheSevenjust or the memory address with 0x40000000, which is the address space limit, so it will wrap around on the memory bus, but not on the protection unit
23:58:48TheSeventhat way we can just circumvent all the caches for these buffers :-D
23:58:59gevaertssimilar to the PP trick

Previous day | Next day