--- Log for 06.01.123 Server: calcium.libera.chat Channel: #rockbox --- Nick: rb-logbot Version: Dancer V4.16 Started: 11 days and 9 hours ago 00.02.43 # Build Server message: 3Build round completed after 1020 seconds. 00.02.46 # Build Server message: 3Revision 9bd5249445 result: 39 errors 0 warnings 00.03.00 # :/ 00.09.34 *** Saving seen data "./dancer.seen" 00.32.27 # uzziyah is back up. 00.33.23 # also btw the debug buildgraphs don't display 00.39.32 # Build Server message: 3New build round started. Revision 8366b8511d, 303 builds, 8 clients. 00.58.09 # Build Server message: 3Build round completed after 1117 seconds. 00.58.11 # Build Server message: 3Revision 8366b8511d result: All green 02.09.38 *** No seen item changed, no save performed. 02.41.05 Quit chris_s (Quit: Connection closed) 04.09.42 *** Saving seen data "./dancer.seen" 04.18.18 Join lebellium [0] (~lebellium@2a01cb040109a600a4e31cf9219b56f3.ipv6.abo.wanadoo.fr) 05.34.41 Join lisa_ [0] (~chatzilla@5.180.61.167) 05.57.22 # Good Morning! Noob here...need help diagnosing the status of my clip+. All it does at start up (using various combos of buttons as described in forum and help sections) is sansa-->flower logo-->sansa-->flower logo. It hangs on flower logo. Not recognized by computer (Linux) through USB. I tried to create a bootable SD card but no luck. I'm open to trying the "shorting" to enter... 05.57.24 # ...recovery mode, but thought I should run the situation by you experts first before performing any invasive feats! Thanks for any morsel you might throw my way. Best regards. 06.09.44 *** No seen item changed, no save performed. 06.35.21 # <_bilgus> lisa do you already have rockbox on this device? 06.38.01 # <_bilgus> Its been a long while but i'm pretty sure you can still get the device to enumerate even with bad flash, you'll need to get that figured out first 06.40.22 # <_bilgus> if you get that to happen and the device isn't read only you should be able to use 'dd' to copy a drive image to the device 06.40.24 # <_bilgus> https://forums.rockbox.org/index.php?topic=52014.0 06.41.54 # <_bilgus> if the device is already read only and you don't have our bootloader already its done save it for parts 06.55.47 # <_bilgus> oh thankfully we have the wiki https://www.rockbox.org/wiki/SansaAMSUnbrick 06.56.14 # <_bilgus> you have to short out the nandto get it to ginto recovery mode 06.59.23 # <_bilgus> and a guide someone made for a different sansa model https://forums.rockbox.org/index.php?topic=29115.0 06.59.36 # <_bilgus> same idea but use the image provided above 07.00.10 # <_bilgus> and i'm pretty certain I never desoldered the battery 07.01.29 # <_bilgus> you just have to make sure its off first 07.07.22 # Hi bilgus, and thanks for the info. Not sure if rockbox was installed since I inherited this device. I will peruse all you've suggested and keep trying. Thanks a bunch. 07.31.04 Quit lebellium (Quit: Leaving) 07.57.13 Join massiveH [0] (~massiveH@2600:4040:a992:a300:80b4:51ad:b14a:7d26) 08.09.46 *** Saving seen data "./dancer.seen" 08.27.30 Quit lisa_ (Quit: ChatZilla 0.14 [SeaMonkey 2.53.14/20220924145453]) 08.44.05 Quit massiveH (Quit: Leaving) 09.37.11 # It would be cool if you could add it back in to show the time, I used it a lot. 09.38.32 # munkis: huh, I don't think I've ever looked at that. 09.45.38 # and it should be fixed now. another PHP8-ism, apparently. 09.51.16 # the debug view does really show just how badly my main builders are suffering due to crappy upload speeds.. 10.09.50 *** Saving seen data "./dancer.seen" 10.23.48 Quit danwellby (Remote host closed the connection) 10.37.27 Join danwellby [0] (~danwellby@cpc1-cdif16-2-0-cust352.5-1.cable.virginm.net) 11.48.10 Join othello7 [0] (~Thunderbi@pool-100-36-133-223.washdc.fios.verizon.net) 12.09.52 *** Saving seen data "./dancer.seen" 12.43.35 Join amachronic [0] (~amachroni@user/amachronic) 12.50.19 # Build Server message: 3New build round started. Revision 5750eb3181, 303 builds, 8 clients. 13.07.23 # Build Server message: 3Build round completed after 1023 seconds. 13.07.26 # Build Server message: 3Revision 5750eb3181 result: All green 13.10.16 # Build Server message: 3New build round started. Revision b444ecfca2, 303 builds, 8 clients. 13.35.08 # Build Server message: 3Build round completed after 1493 seconds. 13.35.10 # Build Server message: 3Revision b444ecfca2 result: All green 13.54.10 # Build Server message: 3New build round started. Revision 5d0c382a59, 303 builds, 8 clients. 14.07.31 # Thanks a lot for the date modified feature. 14.09.53 *** No seen item changed, no save performed. 14.11.26 # Also I managed to get around the calculator.c, mikmod.make and rockaux.c errors with windows simulators, so now I'm building simulators from git without any changes. 14.14.15 # Build Server message: 3Build round completed after 1205 seconds. 14.14.17 # Build Server message: 3Revision 5d0c382a59 result: All green 14.19.30 # Build Server message: 3New build round started. Revision 5bb062391f, 303 builds, 8 clients. 14.41.50 # Build Server message: 3Build round completed after 1340 seconds. 14.41.52 # Build Server message: 3Revision 5bb062391f result: All green 14.44.43 # I still have the voice problem on my gigabeta s though where it says s after each option and it can't say yes or no and it can not spell. Is there a way to solve this? 14.56.58 Join wsa [0] (~wsa@i5E861F60.versanet.de) 14.57.11 # hi! 14.58.41 # I wanted to report that Rockbox currently fails to play files on my Sansa Clip+ (prefetch abort) 14.58.55 # Revert "rbcodec dsp: Refactor DSP init routines, restore INIT_ATTR" fixes the issue 14.59.17 # Sadly, I can't register at flyspray, so I commented in gerrit instead 14.59.26 # https://gerrit.rockbox.org/r/c/rockbox/+/4915 15.00.16 # uh oh that's mine 15.00.24 # I have no experience with the DSP subsystem, so I can't really tell what goes wrong 15.00.46 # but I can apply debug patches or test stuff 15.01.09 # are you able to get a useful traceback? 15.01.44 # on the screen, there is no traceback. is there another way to get one? 15.02.13 # not that I know of 15.02.47 # usually you get lines of pc:xxx but if things are badly broken the traceback can fail 15.03.26 # I'll try again 15.03.40 # i can build locally to see if it's INIT_ATTR causing the problem 15.04.41 # the build doesn't yet check references from non-init to init code so something might be calling an init routine that's been erased. 15.05.04 # with a slightly newer binutils I can do it (and I did so, but I built with another target) 15.06.00 # Prefetch abort at 6772006a 15.06.02 # FSR 0x6a 15.06.14 # (domain 6, fault 10) 15.06.27 # pc:6772006a 15.06.59 # sp:30803b70 15.07.19 # I see 15.08.13 # I need the commit hash to hunt down the traceback 15.08.34 # (I need to build the same version and objdump it) 15.08.38 # 9367ef1 15.08.44 # thanks 15.08.49 # "Fix some Shif related UB -- ASAN" 15.09.18 # I could send you the binary? 15.09.37 # I need the .elf for debug info 15.10.54 # ...and it looks like I didn't build the gcc 8.5.0 toolchain for ARM so I can't do the linker check just yet. 15.11.44 # _bilgus might be able to help he has a clip+ 15.12.04 # so sending my .elf does not help? 15.12.39 # i built the elf locally 15.12.50 # but it seems the address is out of range 15.13.17 # yes, I disassembled the elf and nothing is close to 6772xxxx 15.14.17 # or even 3772xxxx 15.15.30 # you could try stripping INIT_ATTR and INITDATA_ATTR from the dsp library 15.16.27 # okay, with another file, i get another pc: 3332313c 15.16.54 # still doesn't help 15.17.03 # it's probably jumping through a bad function pointer 15.17.24 # wild guess 15.17.48 # I'll try removing the keywords 15.20.48 # that alone does not help 15.21.22 # alright, at least that rules out one class of error 15.34.38 # i'm having a hard time understanding what could be the problem 15.34.38 # i checked the asm code handling dsp_init_fn pointers, but it looks good to me 15.35.17 # maybe I should try to split up the changes into multiple patches? 15.37.30 # if that is possible 15.41.06 # I'll be back in ~30 minutes 15.45.59 # I went back to commit 5883cb4a52 which I had on my player a very long time and that doesn't have the voice problem. I am trying to track it down and will report back. 15.52.24 Quit amachronic (Quit: amachronic) 16.09.55 *** Saving seen data "./dancer.seen" 16.31.15 Quit wsa (Quit: ...) 17.08.21 # <_bilgus> wsa, amachronic i can look t it this eve 17.18.20 # <_bilgus> wsa what kind of file are you playing? I don't get the error with mp3, flac, opus or can you upload a file that fails somewhere? might be metadata related 17.18.39 # <_bilgus> oh they left :/ 18.09.58 *** Saving seen data "./dancer.seen" 18.45.26 Join massiveH [0] (~massiveH@2600:4040:a992:a300:80b4:51ad:b14a:7d26) 18.58.01 # MarcAndersen: "the voice problem" ? Is this unique to the Gigabeast? Did you use the nightly voice stuff or generate your own? 19.25.09 # I recall that the ipod nano 1g also had it, but not sansa c200 or clip zip. It's both the daily build and my own. 19.28.23 # The worst thing is when it spells, it just plays some completely random voice clips. 19.29.32 # /nick Strife89 19.30.18 # odd 19.30.27 Nick saanaito is now known as Strife89 (sid399903@id-399903.lymington.irccloud.com) 19.30.41 # there we go 19.34.53 # I also tried to remove .rockbox completely, install daily + voice with rbutil and reboot with no luck. 20.10.00 *** No seen item changed, no save performed. 20.44.37 Quit massiveH (Quit: Leaving) 20.51.28 Join massiveH [0] (~massiveH@2600:4040:a992:a300:2c8e:f999:b62e:dc5f) 21.21.17 # strange.. 21.21.36 # I meant though, where did the voice files come from? 21.39.47 # <_bilgus> LTNS strife89 21.40.43 # <_bilgus> MarcAndersen, I did a patch to busy wait voice clips might be related any idea about when it started 21.41.00 # <_bilgus> ah nm see you are tracking it 21.42.22 # <_bilgus> g#4800 21.42.25 # 3Gerrit review #4800 at https://gerrit.rockbox.org/r/c/rockbox/+/4800 : 3talk.c add busy loop to talk_spell by William Wilgus 21.44.12 # good to see you, _bilgus :) 21.44.46 # i usually go by the other name now 21.49.15 # <_bilgus> ah got ya 21.49.28 # <_bilgus> I'll try and remember that ;) 21.50.53 # The voice problem seams to be between 5883cb4a52 and f3358eb20a since I tried that build because of the time display in properties 21.51.31 # But that is over a half year worth of commits 21.59.51 # <_bilgus> git bisect excels at that 22.02.29 # <_bilgus> $ git bisect start; git bisect good 5883cb4a52; git bisect bad f3358eb20a; each a separate commend 22.02.32 # <_bilgus> command 22.02.54 # <_bilgus> probably be 10 builds before it finds it 22.03.23 # <_bilgus> what sucks is when its so far back that the build system has changed 22.03.33 # <_bilgus> might try building with plugins off 22.04.10 # _bilgus: should I try or do you? 22.04.26 # you said this happens on the ipodmini? 22.04.39 # <_bilgus> if you can make builds then its going to be quicker for you to 22.04.43 # no nano 1g and gigabeat s I think 22.05.00 # nano1g is the newer samsung core IIRC... 22.05.09 # <_bilgus> turning off plugins makes it quite a bit faster 22.05.10 # And c200 and clip zip works 22.05.10 # gigabeast is PP though IIRC? 22.05.36 # sorry, portalplayer aka the same SoC that the (older) ipods use. 22.06.57 # I will be back tomorrow, and see is there's something new, and maybe compile some tests. 22.06.58 # <_bilgus> tools/configure then 'a' for advanced and 'o' to omit plugins 22.07.27 # cool 22.07.50 # <_bilgus> then n c\for normal and enter 22.08.04 # <_bilgus> sorry 'n' for normal and press enter 22.08.28 # I also need v 22.10.03 *** Saving seen data "./dancer.seen" 23.37.10 Quit m01 (Quit: Konversation terminated.) 23.37.23 Quit massiveH (Quit: Leaving) 23.39.29 Join m01 [0] (~quassel@vps-b172b88b.vps.ovh.net)