Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
In case others are interested, I noticed that the nightly build here: http://downloads.slimdevices.com/nightly/7.9/sc/12c414e/logitechmediaserver_7.9.1~1511211491_all.deb now has binary support for AARCH64 when used with Perl Version: 5.24.1 - aarch64-linux-gnu-thread-multi. Logitech Media Server Version: 7.9.1 - 1511211491 @ Mon Nov 20 21:08:22 UTC 2017 Hostname: rock64 Server IP Address: 192.168.0.100 Server HTTP Port Number: 9000 Operating system: Debian - EN - utf8 Platform Architecture: aarch64-linux Perl Version: 5.24.1 - aarch64-linux-gnu-thread-multi Audio::Scan: 0.95 Database Version: DBD::SQLite 1.34_01 (sqlite 3.7.7.1) Total Players Recognized: 5 This suits Debian "stretch" and Ubuntu "zesty" (17.04). I was able to uninstall my self-built .deb packages and install this one from Michael with no issues on a rock64 SBC running stretch-minimal-rock64-0.5.10-118-arm64.img from 'here' (https://github.com/ayufan-rock64/linux-build/releases/tag/0.5.10) -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
atrocity wrote: > Might this be why DSDplayer and IckStream don't work on an Odroid C2? I found out that I needed to compile dsdplay for aarch64, and copy it into /usr/share/squeezeboxserver/Bin. This would likely work for you too. Code: SUDO APT-GET INSTALL LIBFLAC-DEV SUDO APT-GET INSTALL LIBSOXR-DEV GIT CLONE HTTPS://GITHUB.COM/SQUEEZEONARCH/DSDPLAY.GIT CD DSDPLAY/SRC MAKE CD BUILD SUDO CP ./DSDPLAY /USR/SHARE/SQUEEZEBOXSERVER/BIN/ Getting the Ickstream helper executables to build is harder (at least I couldn't find the source files on Github to give it a try). However, you might be able to get the armhf binaries to work by installing compatibility libraries, such as: Code: SUDO DPKG --ADD-ARCHITECTURE ARMHF SUDO APT-GET UPDATE SUDO APT-GET INSTALL LIBC6:ARMHF LIBNCURSES5:ARMHF LIBSTDC++6:ARMHF The following commands can be used to check whether armhf is currently supported as a foreign architecture and whether the Ickstream armhf binary is loadable. Code: DPKG --PRINT-FOREIGN-ARCHITECTURES armhf CD /VAR/LIB/SQUEEZEBOXSERVER/CACHE/INSTALLEDPLUGINS/PLUGINS/ICKSTREAMPLUGIN *./ickHttpSqueezeboxPlayerDaemon-arm-linux-gnueabihf* Usage: ./ickHttpSqueezeboxPlayerDaemon-arm-linux-gnueabihf IP-address daemonPort wrapperURL logFile authorizationHeader If the above works, i.e., the usage line gets printed out as shown, you are good to use IckStream on your Odroid C2. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
mherger wrote: > > Currently, as my instructions earlier in this thread show, I am > copying > > binaries into /usr/share/squeezeboxserver/Bin/. > > What about the perl Modules? > > > If you would be willing to consider a pull request, I am currently in > > the process of trying to make the existing build scripts run on > aarch64, > > but with a "do no harm" philosophy to ensure all other platforms keep > > building as they do today. See for example: 'here' > > > (https://github.com/crazzell/slimserver-vendor/commit/a381441dde084bdddce360a13e4040d7a156e64e). > > Nice! Did you base this on that other pull request which already > exists? > > I did merge yet another pull request earlier today which should handle > new perl versions much better than before. Make sure you rebase your > changes. > > Michael I was not smart enough to check back here before making my pull request. It did say "This branch has no conflicts with the base branch," so I thought I was okay. Tell me if I need to rebase. Hopefully, this change is pretty simple and shouldn't be hard to merge, but let me know. Best regards, Charles. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
-ol- wrote: > Hi, > > I was wondering whether the described installation procedure could also > work with an Odroid C2 using DietPi as OS (based on Debian Jessie)? > DietPi has its own installation routine for the LMS but the version it > is installing is completely outdated, updates are not possible, too. > Transcoding options (like MP3, Flac, Sox) are not working either. > > Any help would be very much appreciated. > > Thanks & Cheers, > Oliver IIRC, DietPi is still on Debian 8 for the odroidc2, which means it uses automake-1.14, whereas my patch file assumed automake-1.15. You can verify this by typing 'dpkg -L automake | grep guess' to see if "/usr/share/automake-1.14/config.guess" shows up. You can patch the patch as follows: Code: wget -O slimserver-vendor.patch http://forums.slimdevices.com/attachment.php?attachmentid=23873 sed -i 's/automake-1.15/automake-1.14/g' slimserver-vendor.patch If you do as shown above, I give it a reasonable chance of working. Or you can wait a little bit and I suspect Michael will have fixed the main git source repository so that patches aren't needed at all. Charles. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
mherger wrote: > >> But assuming we had binaries for your platform, would the regular > .deb > >> package work? What platform/architecture does Settings/Information > >> report? > > Yes, regular .deb packages would work very nicely, and I would prefer > to > > use them if they were available, rather than brew my own. LMS reports > > "Platform Architecture: aarch64-linux". > > Is there any de-facto standard distribution most people are using? Or > are the most popular options at least using the same Perl version? > Otherwise adding support for this platform would be opening another can > > of worms... > Yes, I see that doing this for multiple perl versions would be painful. I suggest considering Debian Stable (stretch), which is using Perl 5.024, and possibly also Ubuntu 16.04 LTS which is on Perl 5.022. mherger wrote: > > BTW: where do you put the binary files so LMS does find it? It currently > > doesn't support the aarch64 architecture string, does it? > > -- > > Michael Currently, as my instructions earlier in this thread show, I am copying binaries into /usr/share/squeezeboxserver/Bin/. If you would be willing to consider a pull request, I am currently in the process of trying to make the existing build scripts run on aarch64, but with a "do no harm" philosophy to ensure all other platforms keep building as they do today. See for example: 'here' (https://github.com/crazzell/slimserver-vendor/commit/a381441dde084bdddce360a13e4040d7a156e64e). Notice the function refresh_config takes no action at unless the platform is Linux and the ARCH is aarch64. In that sense it is hopefully safe, but your review comments would be welcome. I am doing the same for the other buildme-linux scripts as well (work in progress). Charles. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
Roland0 wrote: > As 'mentioned' > (http://forums.slimdevices.com/showthread.php?108166-Building-LMS-for-64-bit-Debian-running-on-ARM-Single-Board-Computers=897624=1#post897624) > that's how my script does it: > > Code: > > > function refresh_config { > [ -f /tmp/config.guess ] || wget -O /tmp/config.guess 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD' > [ -f /tmp/config.sub ] || wget -O /tmp/config.sub 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD' > cp -vf /tmp/config.guess . > cp -vf /tmp/config.sub . > } > > > > > > If you wanted to check for arch, then one would just put the test into > the refresh_config function. Same approach could be used to check for > availability of wget/curl/network. > > Except for the caveats already mentioned by you (which I think are > minor - wget is available everywhere and usually already installed and > if there is not net connection, one couldn't download the > modules/script in any case), no. Thanks, Roland, you make perfect sense and are well ahead of me as usual. Wouldn't it be nice to get these changes (or something very like them) merged into the main Logitech slimserver-vendor git repo? What were discussing seems like reasonable fix and could close this 'issue' (https://github.com/Logitech/slimserver-vendor/issues/9). -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
mherger wrote: > > I've done all this by hand > > before on different machines, but I like the automation and system > > accounting that Debian packages always provide. > > Ah, the "on different machines" part makes sense. For a single machine > it wouldn't imho. Yes, I had to manually configure things and set up the > > startup script on my server. But updates now only take me a few seconds > > to "git pull" and restart :-). > Okay, I am aware of that usage model -- very nice when tracking nightly changes. mherger wrote: > > But assuming we had binaries for your platform, would the regular .deb > package work? What platform/architecture does Settings/Information > report? > Michael Yes, regular .deb packages would work very nicely, and I would prefer to use them if they were available, rather than brew my own. LMS reports "Platform Architecture: aarch64-linux". -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
mherger wrote: > > But may I ask why you go all the way as to build a .deb file? Why not > just run from the source? > > -- > > Michael Ah, good question.The automation of the Debian package installer is helpful in that it automatically creates /etc/init.d/ scripts, creates the user squeezeboxserver, changes ownership of files so that they can run under that user, and provides proper daemon monitoring (and restarting behavior, if it ever crashes). I've done all this by hand before on different machines, but I like the automation and system accounting that Debian packages always provide. I guess I've become used to the convenience of easily installing and updating Debian packages, and wanted to maintain that convenience, even if I couldn't use your nightly builds anymore due to lack of direct aarch64 support. Charles. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
atrocity wrote: > Great stuff, thanks so much! A couple things, though: > > 1. Am I mis-remembering, or does git also have to be explicitly > installed? > On the armbian version of Ubuntu I picked up to try this, it was already installed. atrocity wrote: > > 2. When I attempted to put the steps for the helper binaries into a > script, the wavpack step failed. I believe the initial cd step should > just be > Code: > > cd ../wavpack > > > Looking at this again, I find an extra pair of dots would be necessary, since you should be in ./vendor/faad2/faad2/frontend at that point. I should have written "cd ../../../wavpack/" atrocity wrote: > > Would running the entire process from the top be the correct way to > update to the latest nightly? > If you have already run this once, and can keep the build directories in-tact (i.e., you have room for them), then updating to the latest nightly should start with the "git pull" commands and go from there. atrocity wrote: > > Also curious if the same steps would work on a C2 running Debian. It's > not my computer and not in my house, so I don't know the exact OS > details and understand if there's no way you can answer that. Just > wondering if I could script all this stuff and send it to a friend. > I had previously developed this process on Debian Stretch, (armbian). I have successfully used it on Debian Jessie too. I only tried it on Ubuntu so I could give you meaningful guidance. atrocity wrote: > > Thank you again for doing this! I was a programmer for decades but my > specialty was SAS and JCL on the mainframe. So I find myself in the > annoying position of having a head full of lots computer/programming > knowledge with almost none of it being relevant here. I was glad to share it, and happy it worked for you. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
Roland0 wrote: > It's a mix of both - some packages do not support aarch64, even in their > newest version. For those, config.guess is updated. > > > > ad 1: automake doesn't seem to be fully downwards compatible, so there > is a risk that it will fail (on my gentoo system, there are 3 versions > of automake installed for that reason...) > ad 2: will only work on debian-based linux variants Agreed, it is not a good idea to do something that is Debian-specific, so that was why I was not totally happy and asked for a more elegant solution. What I really meant was a more robust solution. Another way to copy in the latest, known config.guess file without risking a broken automake would be to use: wget -O config.guess 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD' or curl 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD' > config.guess The only assumptions used here a working wget (or curl) and an Internet connection. These extra build steps would be protected by an if [ "`uname -m`" == "aarch64" ]; then conditional, so the probability of breaking the script for existing platforms for which it works would be minimal. Any thoughts on why the above is not a good change to propose? Charles. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
atrocity wrote: > Your hints were perfect. I started the process last night and found it > had finished when I woke up this morning. I installed everything via > sudo dpkg -i with the only snag being that LMS complained about the CPAN > package not being installed. So I installed it along with the two > others, then re-installed LMS, which is now cheerfully scanning my > 100,000+ tracks. > > THANK YOU! You're welcome! Just so that this thread can be more useful to any others who want to follow this method, I retraced my steps from a new install of Ubuntu 16.04.3 LTS on an odroidc2 to LMS installed and running. Here they are: Code: sudo apt-get update sudo apt-get upgrade sudo apt-get install fakeroot sudo apt-get install yasm sudo apt-get install debhelper quilt libz-dev libgd-dev sudo apt-get install libmodule-build-perl sudo apt-get install libio-socket-ssl-perl sudo apt-get install libmodule-install-perl git clone --recursive -b public/7.9 https://github.com/Uplink03/logitechmediaserver-deb.git cd logitechmediaserver-deb/source/vendor/ git pull https://github.com/Logitech/slimserver-vendor.git cd ../server/ git pull https://github.com/Logitech/slimserver.git cd ../platforms/ git pull https://github.com/Logitech/slimserver-platforms.git cd ../vendor/ wget http://forums.slimdevices.com/attachment.php?attachmentid=23873=1508605856 slimserver-vendor.patch mv attachment.php\?attachmentid\=23873 slimserver-vendor.patch patch -p2 < slimserver-vendor.patch rm slimserver-vendor.patch cd ../.. dpkg-buildpackage -rfakeroot -b -us -uc cd ls -l *.deb sudo dpkg -i logi*.deb sudo /etc/init.d/logitechmediaserver start systemctl status logitechmediaserver.service However, you may want the specially-patched versions of the helper binaries for flac, sox and faad as well. Code: cd ./logitechmediaserver-deb/source/vendor/flac/ ./buildme-linux.sh tar xvf flac-build-aarch64-.tgz --strip-components=7 sudo cp ./bin/flac /usr/share/squeezeboxserver/Bin cd ../sox ./buildme-linux.sh tar xvf sox-build-aarch64-.tgz --strip-components=7 sudo cp ./bin/sox /usr/share/squeezeboxserver/Bin/ cd ../alac_decoder make sudo cp ./alac /usr/share/squeezeboxserver/Bin/ cd ../faad2 git clone https://github.com/ralph-irving/faad2.git cd faad2/ cp `dpkg -L automake | grep guess` . ./configure make cd frontend/ sudo cp ./faad /usr/share/squeezeboxserver/Bin/ cd ../../wavpack sed -i '/cd wavpack-4.50.1/a cp `dpkg -L automake | grep guess` .' build.sh ./build.sh sudo cp ./wvunpack /usr/share/squeezeboxserver/Bin/ Hopefully this helps someone. Charles. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
atrocity wrote: > Thank you! There's definitely no hurry, though. I don't strictly need > another server, it's just fun to play around with this stuff. I got as > far as running git apply --check on the patch and wound up with: > > > Code: > > > error: slimserver-vendor-patched/CPAN/buildme.sh: No such file or directory > error: slimserver-vendor-patched/flac/buildme-linux.sh: No such file or directory > error: slimserver-vendor-patched/sox/buildme-linux.sh: No such file or directory > > > > Thank you again! 1) Make sure that /sbin is in your PATH. I needed to edit my ~/.profile to include PATH=$PATH:/sbin:/usr/sbin 2) When applying the patch, do so as follows: Code: cd ./logitechmediaserver-deb/source/vendor patch -p2 < ~/slimserver-vendor.patch -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
Roland0 wrote: > See 'this github issue' > (https://github.com/Logitech/slimserver-vendor/issues/9) for > discussion. > See 'this post' > (http://forums.slimdevices.com/showthread.php?107222-Howto-build-LMS-for-aarch64) > for a slightly more elegant modification for the the build script (as is > fetches the newest config.guess automatically and doesn't depend on a > local copy), plus a update to build the newest versions of the libraries > / perl modules (a LMS built this way has been running on my 64bit RPi3 > for a couple of months without any issue). Okay, good to know this has been tackled by others before me. I see the patched buildme.sh updates the troublesome, outdated packages to overcome the staleness of the config.guess. My thought was not to stray further from the beaten path than necessary, so using the same (outdated) source code as the official Logitech build scripts seemed a better idea to me, but that's arguable both ways. I have experimented with actually running automake again instead of just copying in a newer config.guess, e.g., invoke automake -a -f, in all those places where outdated packages won't build. I'm not sure if that's better or worse, but it does work. We could have the following inserted at all the critical places in buildme.sh Code: if [ "`uname -m`" == "aarch64" ]; then automake -a -f fi or Code: if [ "`uname -m`" == "aarch64" ]; then cp `dpkg -L automake | grep 'config.guess'` . fi Either of these would be non-destructive to normal users of buildme.sh. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
atrocity wrote: > Is there a guide out there for dummies who are comfortable with the > command line and have git installed on an Odroid C2 but have never tried > to do something like this from scratch? Or is the first post in this > thread all I need? I can try to post more detailed steps, and/or provide feedback in this thread if you get stuck. I did this on an odroidc2, as well. Are you running Jessie or Stretch and which system Perl is installed? (Also, is this armbian or dietpi or some other variant?) A few things I neglected to mention in the first post were: Code: sudo apt-get install fakeroot sudo apt-get install yasm sudo apt-get install debhelper quilt libz-dev libgd-dev sudo apt-get install libmodule-build-perl sudo apt-get install libmodule-install-perl sudo apt-get install libio-socket-ssl-perl uname -a; perl -v, on my machine shows the following: Code: Linux music2 4.13.7-odroidc2 #32 SMP PREEMPT Mon Oct 16 21:38:07 CEST 2017 aarch64 GNU/Linux This is perl 5, version 24, subversion 1 (v5.24.1) built for aarch64-linux-gnu-thread-multi (with 75 registered patches, see perl -V for more detail) I will try to do better write-up in the next 24 hours, but to make it better targeted to your system please let me know which OS (cat /etc/os-release) and which perl (perl -v) you are starting with. Thanks. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
[SlimDevices: Unix] Building LMS for 64-bit Debian running on ARM Single Board Computers
1) Motivation: I don't want to run a power-hungry PC 24/7 with spinning fans etc, but I want something faster the a Raspberry Pi to run LMS (I still use the web interface). There are number of fast, cheap SBC's out there that handily out-perform the RPi3, such as OdroidC2 (which supports an eMMC module for storage), ROCK64 (which has USB3 and eMMC). Quad-Core ARM Cortex A53 64-Bit Processors and support for 2GB or 4GB of 1600MHz LPDDR3 memory, endow these tiny machines with similar power to the x86 PCs many of us used as LMS servers back in the day. 2) Issue: ARM64/AARCH64 architectures are not currently supported in the highly build Debian packages as of now. It seems you have to roll your own. 3) My solution: Build Debian install packages using the method and scripts supplied here: https://github.com/Uplink03/logitechmediaserver-deb. This, in turn, pulls together the source repositories from https://github.com/Logitech/slimserver-vendor.git, https://github.com/Logitech/slimserver.git and https://github.com/Logitech/slimserver-platforms.git. After checking out the repo Code: git clone --recursive -b public/7.9 https://github.com/Uplink03/logitechmediaserver-deb.git I wanted to make sure I was using fully updated sources, so I did the following (probably not the right way to do it, but it worked for me). Code: cd logitechmediaserver-deb/source/vendor git pull https://github.com/Logitech/slimserver-vendor.git cd ../server git pull https://github.com/Logitech/slimserver.git cd ../platforms/ git pull https://github.com/Logitech/slimserver-platforms.git Now the only problem I has was that those CPAN libraries that use automake/configure generally fail to complete the ./configure script on 64-bit ARM machines. To fix this, I modified ./logitechmediaserver-deb/source/vendor/CPAN/buildme.sh to copy the platform's config.guess file into each of the unarchived source directories, immediately before running ./configure. A suitable patch for doing this is attached, but you might have to edit the path to config.guess to suit your installation,(dpkg -L automake | grep guess) will tell what it should be. In my case, it was /usr/share/automake-1.15/config.guess. This simple trick is enough to allow ./logitechmediaserver-deb/source/vendor/CPAN/buildme.sh to run to completion without further issues. Then, it is sufficient to the following command in ./logitechmediaserver-deb Code: dpkg-buildpackage -rfakeroot -b -us -uc This should leave you with the following files, ready to install with dpkg -i Code: rw-r--r-- 1 charles charles 4476976 Oct 20 21:16 logitechmediaserver_7.9.0+lmce1_all.deb -rw-r--r-- 1 charles charles 2549430 Oct 20 21:16 logitechmediaserver-code2000-font_7.9.0+lmce1_all.deb -rw-r--r-- 1 charles charles 4908774 Oct 20 21:16 logitechmediaserver-cpan-bundle_7.9.0+lmce1_arm64.deb -rw-r--r-- 1 charles charles 1745448 Oct 20 21:15 logitechmediaserver-firmware_7.9.0+lmce1_all.deb -rw-r--r-- 1 charles charles 3396910 Oct 20 21:16 logitechmediaserver-html_7.9.0+lmce1_all.deb This is a reasonably easy route to follow, the only trickery being the need to patch CPAN/buildme.sh to work around automake's config.guess being so old in many packages. Can anyone suggest a more elegant way to solve this, that we could perhaps get merged into the official logitech repo version of buildme.sh? +---+ |Filename: slimserver-vendor.patch | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=23873| +---+ -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=108166 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Pi - on the case?
Man in a van wrote: > https://www.fruitcore.de/produkte/case/hifiberry/ --- Ships to France, > Germany, Austria and the UK only, according to the options listed in their on-line checkout menu wortgefecht wrote: > Well, most of my hifi devices (that are not hooked to my LMS setup and > therefore not mentioned in my footer) are classic silver: Marantz, > Sansui, ReVox, ... [emoji13] > > But there is a black case for DAC+ RCA available: > https://www.pi-shop.ch/metallgehaeuse-fuer-hifiberry-dac-rca-fuer-pi-b-und-pi-2-schwarz > --- Ships to Switzerland and Lichtenstein only, according to the options listed in their on-line checkout menu. wortgefecht wrote: > > > EDIT: Since you live in the UK, maybe it would be cheaper for you to > order it from Belgium: > http://shop.mchobby.be/boitiers/665-boitier-metal-noir-pour-raspberry-et-hifiberry-dac-323216652.html > --- Ships to Belgium, France, Luxembourg & The Netherlands and Germany only (according to their recent message to me, quoted below). > Bonjour, > En effet, nous ne livrons que en Belgique, France, Luxembourg, Pays bas > et Allemagne actuellement. > Les exportations hors Europe ne sont pas dans les prévisions des années > à venir. I think these stores are doing a good job of sticking to their territorial agreements. However, if you go to the main HiFiBerry website, you'll find that black is back (at least for now, and they're willing to ship worldwide as far as I can tell). https://www.hifiberry.com/product/steel-case-for-hifiberry-dac-brushed-cover-2/ Charles. -- "Dreamer, easy in the chair that really fits you..." dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=104666 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Fedora LMS RPM packages, call for testers...
JackOfAll wrote: Charles, I'm being dim. Of course, several cmds are run from jivelite using sudo. They are listed in /etc/sudoers.d/01_jivelite. Update the jivelite package. (sudo sc-cleanUpdate jivelite) Add your user to jivelite group. (sudo usermod -a -G jivelite your_user_name) Then, just the cmds listed in 01_jivelite will be run sudo without asking for password, if you are not happy with global sudo without password for users in wheel group. If currently logged into X session. End it and log back in. Try starting jivelite from the icon. Should work fine. Clive, yes that did it. Thanks! Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101931 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Fedora LMS RPM packages, call for testers...
Code: [crazzell@fedora21 ~]$ jivelite JiveLite 0.1.0 [sudo] password for crazzell: [crazzell@fedora21 ~]$ rpm -q jivelite jivelite-0.1-0.81.20140412gitad694dc.fc22.x86_64 [crazzell@fedora21 ~]$ uname -a Linux fedora21.localdomain 4.1.4-200.fc22.x86_64 #1 SMP Tue Aug 4 03:22:33 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux It would be nice to launch Jivelite from the Gnome 3.16 launcher, but I can't because it's asking for a password on the console frame buffer, which is obscured by the currently running X-windows desktop. Any help would be appreciated. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101931 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] multiple ffmpeg instances
This worked rather easily for me... Code: wget http://ffmpeg.org/releases/ffmpeg-2.5.4.tar.bz2 tar xvf ffmpeg-2.5.4.tar.bz2 cd ffmpeg-2.5.4 ./configure make -j4 # occasionally, having a quad core pays off! apt-get remove ffmpeg sudo make install ps -ef | grep ffmpeg 1001 14092 1 8 11:17 ?00:01:20 /usr/local/bin/ffmpeg -loglevel quiet -i http://a.files.bbci.co.uk/media/live/manifesto/audio/simulcast/hls/nonuk/sbr_low/ak/bbc_radio_three.m3u8 -f flac - Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=103150 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: Squeeze on Arch - developer version
Triode wrote: I've upgrade the install images to improve the speed of installation of SoA. Please let me know how you get on Worked very nicely using Reinstall SoA button. Speed much improved. I originally installed LMS 7.8, changed my mind because it didn't scan DSF files, removed LMS 7.8 and installed 7.9. A couple of things I tend to do that your install scripts omit: 1) Generating and setting the locale, I uncomment en_US.UTF-8 UTF-8 in /etc/locale.gen, and locale-gen run as root to generate it, set as the system-wide default running localectl set-locale LANG=en_US.UTF-8 as root, and reboot to let the settings take effect. 2) Enable and start avahi-daemon using systemctl enable avahi-daemon.service; systemctl start avahi-daemon.service. My motivation for #1 is for LMS to inherit an appropriate locale before doing a scan, so that any tags with accented characters display properly. My motivation for #2 is to use soa-wandboard.local as a hostname rather than the dotted-decimal IP address in my web browser sessions and ssh sessions. Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101624 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: Squeeze on Arch - developer version
JackOfAll wrote: Hi Charles, happy new year! Yep, fuster cluck, a complete fuster cluck. Which was even worse when I looked at the F21 3.17 mainline kernel and they have 2x dt files, one for B1 and another for C1. I did have everything working OK I think, on a 3.16.7, with runtime detection for B1/C1 via the rfkill driver, which isn't in mainline. It was built with RN patchset, plus the attached patches. On my list of things to do, and the only thing delaying me uploading a F21 package set for armv7hnl to the SC repo, sort this out with the F21 3.17 kernel I'll give you a shout when I've done that. Out of interest, do you have both a B1 and C1 revision board? Happy new year to you too, Clive! I do have both B1 and C1 revision boards (both of them Dual Lite versions). I appreciate the patches too, which might just tempt me into spending even more time on this than I have already... :-) Would be glad to do any testing that's a bit more directed than my, let's just try another kernel and cross my fingers approach. Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101624 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: Squeeze on Arch - developer version
Okay, so this is my OCD coming through, but I would really like to find a kernel reliably initializes the Broadcom 4329/4330 chips on the Wandboards and reboots cleanly. I have built and tested quite a few kernels, either cross-compiled according instructions here, https://eewiki.net/display/linuxonarm/Wandboard#Wandboard-LinuxKernel, or natively compiled using Arch's elegant PKGBUILD / makepkg mechanism e.g., from https://github.com/CrawX/PKGBUILDs/tree/wblinux3.14/core/linux-wandboard. In short, nothing I've tried has been fully reliable: either WLAN needs a cold boot, or with the more recent kernels (3.18 or higher) WLAN comes on about 50% of warm boots for me. I've given up this fool's errand for now :(, but plan to keep on using 3.14.14-1 from CrawX's PKGBUILD until something better comes along. Robert Nelson has told me that dt-bindings for brcmfmac is still in the bike-shed painting phase, which I take to mean the people are still arguing about how this should be done rather than actually doing it. If someone happens to have more insight on this issue, I'd appreciate hearing about it.:) -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101624 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: Squeeze on Arch - developer version
ian_heys wrote: I've had this minor inconvenience since I first installed three or four months ago (Wandboard Quad Rev C). I use the reset button during install phases and reboot doesn't work from the UI. Installed irqbalance and I still have it. Ian, thanks for that, I tried the same thing and made sure to enable irqbalance after installing it: Code: pacman -S irqbalance systemctl enable irqbalance.service systemctl start irqbalance.service Anyway, this didn't help at all. I consider it a minor inconvenience as well, but I do think that new users could get tripped up by during the installation process. Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101624 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: Squeeze on Arch - developer version
ian_heys wrote: I can't test at the moment as I don't have a cable. I do Putty/SSH into the wandboard but over the network. Do I need a straight or null-modem cable to connect to my PC for testing? My long term plan for the WB is to use it headless in a bedroom, so no long term potential to have the serial port connected. Just to chime in here: I have been testing this intensively with a serial cable attached to view where it gets stuck, and I have my case as evidence that if you have this problem without a serial cable, it doesn't get any better with a serial cable attached :-) This problem has been discussed in other fora, e.g., http://archlinuxarm.org/forum/viewtopic.php?f=45t=7586, but even though user hansan seems to have made progress, I don't see anything shared back with the community that I could use. Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101624 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: Squeeze on Arch - developer version
Over Christmas, I decided to try out Triode's SOA on a Wandboard Dual Lite (rev b). Everything went smoothly, with the small exception that the Arch kernel refuses to warm boot on my hardware. I had to pull the plug several times to get SOA installed. I tried updating the kernel to the modified one and also tried replacing the u-boot with one I compiled from sources. All permutations resulted in a working system when booting from a power-on reset, but no success in booting when using sync; reboot from the command line. Later versions of the Linux kernel seem to have no such issues (3.18.1-arm7 for example), but I am having trouble reliably initializing the Broadcom WiFi chip with the newer kernel. I see that this has been discussed earlier in this thread and someone thought that installing irqbalance solved it. Others said the problem improved over time, but I never saw a complete resolution. Is this problem hardware-specific to particular versions of the Wandboard? Overall, I congratulate Triode for making SOA so complete and user-friendly! -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101624 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] any issues with LMS 7.8.x or 7.9.x on Debian Wheezy (x86)?
adamdea wrote: Thanks -this is very interesting. Apologies if I am asking dumb questions (I am very new to linux) but is the idea that one can execute by a)simply executing the commands in your first post line by line. -Where do the patches need to be? b) then executing the script in your second post Is there a way of turning your build into a .deb package as I notice that someone did with the 64bit perl 5.20 version of LMS That said the standard debian image used for Cubox-i is jessie and I think it may be using 5.18 I think the image was made in July and so hasn't gone to 5.20, but I suspect it won;t work with 5.14 You can execute the scripts from both of my posts line-by-line, or (better idea) copy and paste the whole code: segment from each post into its individual file and execute as a script. I have attached shell script versions as files to make it easier to try. (Source them by putting a dot in front on the commands line such as . build_cpan.txt and, if successful, then also use . add_cpan.txt). I had to make them .txt files in order to attach them to this post. The patch files are expected to be in same directory the one from where the scripts are executed. There is a way to make a .deb package, but I haven't had time to do that and I currently only have a build for my immediate need which was Perl 5.14 on Wheezy. +---+ |Filename: build_cpan.txt | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=16406| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=102075 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] any issues with LMS 7.8.x or 7.9.x on Debian Wheezy (x86)?
adamdea wrote: Yes it did work, thanks. Mind you I am trying to repeat the exercise using Debian on a cubox-i and finding it very hard going. But that's getting LMS running I haven't got to squeezelite yet Have you had a go Now you mention cubox-i, which is i.MX6 based, it doesn't seem out of place to offer my build ideas for Wandboard running Wheezy, which has the same SoC at its core. I build my CPAN modules as follows: Code: #!/bin/bash if [ ! -d ./slimserver-vendor ] then git clone https://github.com/Logitech/slimserver-vendor.git fi if [ ! -f YAML-LibYAML-0.52.tar.gz ] then wget http://search.cpan.org/CPAN/authors/id/I/IN/INGY/YAML-LibYAML-0.52.tar.gz fi if [ ! -f Compress-Raw-Zlib-2.065.tar.gz ] then wget http://search.cpan.org/CPAN/authors/id/P/PM/PMQS/Compress-Raw-Zlib-2.065.tar.gz fi if [ ! -f Audio-Scan-0.95.tar.gz ] then git clone https://github.com/SqueezeOnArch/Audio-Scan.git mv ./Audio-Scan ./Audio-Scan-0.95 tar cf Audio-Scan-0.95.tar ./Audio-Scan-0.95/ rm -rf ./Audio-Scan-0.95 gzip Audio-Scan-0.95.tar fi cp Audio-Scan-0.95.tar.gz ./slimserver-vendor/CPAN cp Compress-Raw-Zlib-2.065.tar.gz ./slimserver-vendor/CPAN cp YAML-LibYAML-0.52.tar.gz ./slimserver-vendor/CPAN cp Font-FreeType-0.03.patch ./slimserver-vendor/CPAN cp buildme.patch ./slimserver-vendor/CPAN cp Image-Scale-0.08.patch ./slimserver-vendor/CPAN cd ./slimserver-vendor/CPAN patch buildme.sh buildme.patch PERL_514=/usr/bin/perl ./buildme.sh The 3 patches needed are attached to this post. +---+ |Filename: Font-FreeType-0.03.patch | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=16378| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=102075 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] any issues with LMS 7.8.x or 7.9.x on Debian Wheezy (x86)?
The following script must be run in the same directory as was used for the CPAN build script above. It depends on having a ./slimserver-vendor directory with a successfully built arm-linux-gnueabihf-thread-multi-64int shared object file set. Code: #!/bin/bash thisdir=`pwd` if [ ! -f ./logitechmediaserver-7.8.0-noCPAN.tgz ] then wget http://downloads.slimdevices.com/LogitechMediaServer_v7.8.0/logitechmediaserver-7.8.0-noCPAN.tgz fi if [ ! -d ./logitechmediaserver-7.8.0-noCPAN ] then tar xf logitechmediaserver-7.8.0-noCPAN.tgz fi cd logitechmediaserver-7.8.0-noCPAN/CPAN/ cp -r $thisdir/slimserver-vendor/CPAN/build/5.14/lib/perl5/arm-linux-gnueabihf-thread-multi-64int/* . cd ./arch cp -r $thisdir/slimserver-vendor/CPAN/build/arch/* . cd .. rm -rf ./auto cd .. patch -p1 $thisdir/0001-lms-server-DSD.patch patch -p1 $thisdir/0002-lms-server-aiff-non-transcode.patch patch modules.conf $thisdir/modules.patch cd $thisdir git clone https://github.com/SqueezeOnArch/dsdplay.git cd dsdplay/ cd src/ make cd build/ cp -v dsdplay $thisdir/logitechmediaserver-7.8.0-noCPAN/Bin cp -v $thisdir/dsdplay/src/doc/custom-convert.conf $thisdir/logitechmediaserver-7.8.0-noCPAN/ The attached patches are intended to enable DSD support, and allow LMS to use the Audio-Scan-0.95 version that results from the build procedure I outlined in the previous post. +---+ |Filename: 0002-lms-server-aiff-non-transcode.patch | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=16383| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=102075 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] any issues with LMS 7.8.x or 7.9.x on Debian Wheezy (x86)?
I know you asked about x86, but I don't see why it should be much different from my results on arm7hf Code: debian@arm:~$ cat /etc/os-release | awk 'NR==1 {print $0}' PRETTY_NAME=Debian GNU/Linux 7 (wheezy) debian@arm:~$ uname -a Linux arm 3.13.6-armv7-x11 #1 SMP Sat Mar 8 16:45:58 PST 2014 armv7l GNU/Linux debian@arm:~$ uptime 20:35:15 up 39 days, 9:18, 1 user, load average: 0.00, 0.02, 0.05 And from Logitech Media Server's information page: Code: Logitech Media Server Version: 7.8.0 - 1394196562 @ Sat Mar 8 19:18:16 PST 2014 Hostname: arm Server IP Address: 10.0.1.202 Server HTTP Port Number: 9000 Operating system: Debian - EN - utf8 Platform Architecture: armv7l-linux Perl Version: 5.14.2 - arm-linux-gnueabihf-thread-multi-64int Database Version: DBD::SQLite 1.34_01 (sqlite 3.7.7.1) Total Players Recognized: 5 So, no issues at all, stable as a rock over the past 6 months using the stock Perl (5.14.2). -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=102075 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Fedora LMS RPM packages, call for testers...
Thanks for putting these repos up. I had been using Debian 7 on my Wandboards for the the past few months, but the availability of repos for F20 makes it viable to reconsider that decision. I started with the minimal F20 image from here: http://download.fedoraproject.org/pub/fedora/linux/updates/20/Images/armhfp/Fedora-Minimal-armhfp-20-20140407-sda.raw.xz, followed by Code: sudo yum update cd /boot rm dtb ln -sf dtb-3.15.10-201.fc20.armv7hl dtb sudo rpm -Uhv http://www.squeezecommunity.org/repo/20/armhfp/squeezecommunity-repo-1-1.noarch.rpm sudo yum install squeezelite I haven't gone further than that, but nice that I could get to listening to music so quickly and easily. Dependency resolution is wonderful thing when you've been living without it for a while... Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=101931 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Last night I was mulling over the LMS versions again.. and thinking OK, I'll pull the 7.8.0 release tag from git, make that the stable version on F19 and go forward with 7.8.1 on F20. But from my point of view, unless you guys can actually give me reasons not to make 7.8.1 the default across all versions and platforms, I haven't got any real reason not to do that. The stuff that was broken, as far as I can see, is now working. I would rather that only those who venture into the testing repos, and thereby indicate their willingness to live on the bleeding edge get the 7.8.1 versions of LMS. I would make this the rule regardless of whether we are talking F19, F20, ARM, or Intel. My reasons: - What is broken or fixed on any given day will change with 7.8.1 - Giving people choices is usually good. - -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: The issue here, is that there isn't a 7.8.0 branch. public/7.8 has a tagged 7.8.0 build and now continues. Any fixes to 7.8.0 are going to go into public/7.8, which means at the point we need to fix 7.8.0 we will be shipping a 7.8.x 7.8.0 (ie. 7.8.1, 7.8.2) in stable to do so. I see your point. The usual best practice of back-porting important fixes from the development trunk into the previous version release branch can't happen because there is no branch. The only option that would leave would be to have a set of CSOS-specific patches to 7.8.0 that get applied when building the CSOS RPM, but without changing the version number advertised to the user. This is what you had been doing anyway to patch DSD support into 7.8.0, which implies it is feasible, although I can see several reasons why it may not be desirable. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: I'm in a teasing mood.. Let's see if some of our original users, (who seem to heading off in a different direction now), are still reading this thread.. [image: http://www.communitysqueeze.org/logos/beagle-web-gui.png] Code: [root@beaglebone ~]# rpm -qa | grep botic | sort -d akmod-botic-0.3-2.fc20.armv7hl botic-kmod-common-1.0-1.fc20.noarch kmod-botic-3.13.7-200.1.bone.fc20.armv7hl-0.3-2.fc20.armv7hl [root@beaglebone ~]# rpm -ql kmod-botic-3.13.7-200.1.bone.fc20.armv7hl-0.3-2.fc20.armv7hl /usr/lib/modules/3.13.7-200.1.bone.fc20.armv7hl/extra /usr/lib/modules/3.13.7-200.1.bone.fc20.armv7hl/extra/botic /usr/lib/modules/3.13.7-200.1.bone.fc20.armv7hl/extra/botic/snd-codec-es9018.ko /usr/lib/modules/3.13.7-200.1.bone.fc20.armv7hl/extra/botic/snd-davinci-botic.ko FC20 on Beaglebone (Black?) with some interesting DAC support in kernel modules running under the current stable 3.13.17 Linux kernel. Looks promising, but for a different project than CSP. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
dsdreamer wrote: I can't get back to this until mid-week (Wednesday evening), much as I want to. Code: [20:43:49.660326] output_init_alsa:636 init output [20:43:49.661168] output_init_alsa:658 requested alsa_buffer: 40 alsa_period: 4 format: 24_3LE mmap: 0 [20:43:49.661291] output_init_common:334 outputbuf size: 3528000 [20:43:49.668433] output_init_common:394 supported rates: 48000 44100 32000 [20:43:49.769650] output_init_alsa:670 memory locked [20:43:49.770703] output_thread:465 open output device: hw:CARD=DWHDMI [20:43:49.770738] output_init_alsa:693 set output sched fifo rt: 45 [20:43:49.772069] alsa_open:234 opened device hw:CARD=DWHDMI using format: S24_LE sample rate: 44100 mmap: 0 [20:43:49.772287] alsa_open:313 buffer: 40 period: 4 - buffer size: 1764 period size: 441 [20:43:49.827557] output_flush:407 flush output buffer [20:43:49.827807] output_flush:407 flush output buffer [20:43:53.321462] output_flush:407 flush output buffer [20:43:53.691921] _output_frames:59 start buffer frames: 32768 [20:43:53.692146] _output_frames:140 track start sample rate: 44100 replay_gain: 0 Still runs too fast, sounds like 45rpm playback of a 33rpm record. Code: [root@localhost sub0]# cat status state: RUNNING owner_pid : 543 trigger_time: 238.716831574 tstamp : 552.649560757 delay : 1764 avail : 0 avail_max : 441 - hw_ptr : 16779168 appl_ptr: 16780932 Judging by the rate at which the apple_ptr is advancing timed using a stop watch, I think the playback sample rate is about 54kHz on 44.1kHz files. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: I'm not expecting to see 352k8 and 384k in the supported sample rates. I'm not looking at the code, but I thought max was 192k Anyway, neither here nor there at the moment. What I would like to see is you accessing the hw device directly, not using default. Can you do this again with hw:CARD=DWHDMI so you are accessing the device directly and removing any alsa magic of whatever is lurking behind the default recipe, please. I can't get back to this until mid-week (Wednesday evening), much as I want to. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
ab.wagener wrote: Hi Clive, I tested the kernel and for the first time a HDMI device was visible. fbset shows the correct geometry of the attached screen, but all timings are set to null. The hdmi audio device is also present. My physical screen has two dhmi inputs. If I switch the screen to the wandboard , I get the error message : Input Signal Out of Range and I'm not able to switch back to my pc-system. I have to ssh from my notebook, shutdown the wandboard, the I have access to my pc again. Also : half succes, but not complete. Bernd I tried this kernel and display came on showing a login prompt (I have Jivelite disabled atm). The resolution was 1024x768 with no kernel parametes, although the max resolution support is 1920x1080p. I can force the resolution to via a video= line to match the monitor. I have been messing around trying to get sound out and did briefly succeed, but with a speed-up effect. It might have been trying to sample 44.1 at 48kHz, but it was defnitely playing too fast. I haven't been able to isolate why I sometimes get sound and other times get silence when attempting to play from squeezelite. I examined the content of cat /proc/asound/card2/pcm0p/sub0/status, which is the card corresponding to DWHDMI and I could see that the interface was running and the pointers were advancing as expected, even when no sound audible. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Thanks Bernd and Charles. I've been out all day so still haven't started looking at this myself. Maybe tomorrow Thoughts Some progress. I'm thinking that the EDID resolution is a bit hit and miss, so not surprised that a kernel video= is required. I'm also thinking that the same may apply to audio. There is also some detection code there, trying to get the supported sample rates from the device. Without wishing to cast any doubt in the direction of the codes author, I don't think this code has seen much usage, if any, by anyone other than the author, so I'm not going to be surprised if there are issues. (To put it politely, I doubt there has been any wide testing on a variety of devices. I don't even think Robert Nelson has included it in any of his WB patch sets.) I think that we need to start by enabling debugging in squeezelite, that will show the sample rates that the hdmi audio device claims to support. I think it might be good, to disable MMAP from squeezelite ALSA params, and set format to S24_3LE. Charles, have you had any debugging enabled for squeezelite while trying to play audio over the hdmi device? EDIT: Charles, you might also try -r 48000 -u X, so 44k1 is re-sampled to 48k. Does that play at the correct speed? First attempt is with sample rate forced to 48kHz, by using r 48000 -u X, per your suggestion. Music played, but too fast again. A second attempt at 44.1kHz is also logged below. Code: [21:52:51.303018] output_flush:407 flush output buffer [21:52:51.660772] _output_frames:59 start buffer frames: 8029 [21:52:51.661014] _output_frames:140 track start sample rate: 48000 replay_gain: 0 [21:52:51.671170] output_thread:465 open output device: default:CARD=DWHDMI [21:52:51.674171] alsa_open:234 opened device default:CARD=DWHDMI using format: S32_LE sample rate: 48000 mmap: 0 [21:52:51.674997] alsa_open:313 buffer: 40 period: 4 - buffer size: 1920 period size: 480 [21:55:24.026542] output_close_alsa:698 close output [21:55:24.148493] output_init_alsa:636 init output [21:55:24.149263] output_init_alsa:658 requested alsa_buffer: 40 alsa_period: 4 format: 24_3LE mmap: 0 [21:55:24.149383] output_init_common:334 outputbuf size: 3528000 [21:55:24.173861] output_init_common:394 supported rates: 384000 352800 192000 176400 96000 88200 48000 44100 32000 24000 22500 16000 12000 11025 8000 [21:55:24.264457] output_init_alsa:670 memory locked [21:55:24.265592] output_thread:465 open output device: default:CARD=DWHDMI [21:55:24.265688] output_init_alsa:693 set output sched fifo rt: 45 [21:55:24.268752] alsa_open:234 opened device default:CARD=DWHDMI using format: S32_LE sample rate: 44100 mmap: 0 [21:55:24.269576] alsa_open:313 buffer: 40 period: 4 - buffer size: 1764 period size: 441 [21:55:24.311222] output_flush:407 flush output buffer [21:55:24.313106] output_flush:407 flush output buffer [21:55:38.211258] output_flush:407 flush output buffer [21:55:38.595033] _output_frames:59 start buffer frames: 4096 [21:55:38.595242] _output_frames:140 track start sample rate: 44100 replay_gain: 0 I hope this helps. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: dsdreamer (Charles), do you use hdmi audio? If you do, can you test a F20 kernel? Code: sudo csos-cleanUpdate-testing kernel It's 3.14.0-0.rc4 plus RN patch set, plus the rather large RK imx-drm patch from greghk's staging-next, plus 2 RK patches to add support for imx hdmi audio and a hdmi CEC module, which I added for completeness, as it was a part of his original preview series of patches but not part of the staging-next patch. (Which applies to the hdmi audio driver as well. That didn't become part of the submitted staging-next patch, so perhaps it means that he isn't comfortable that it is ready for prime time usage.) I haven't tested this other than to say it boots. (I've had very limited time to do anything CS related for the last 2 days and I don't have time to do anything more right now than type this message.) The imx fb is no longer a stand-alone module and looks like it wont be created/initialized without a hdmi device being plugged in, which is a PITA from a jivelite running as a service, point of view. Also, I don't believe the hdmi audio device will be created without a hdmi device being detected. (Looks like there is a bunch of initialisation code, that parses ELD for supported sample rates, from the actual plugged-in device, but I've not really looked too closely.) Questions are With a hdmi device being plugged in from boot. 1. Video still works? 2. Does it work any better than it did before, vis-a-vis EDID detection code. (ISTR you had to set params on kernel cmd line.) 3. Does /dev/fb0 get created? 4. Do you get a hdmi audio device listed? (aplay -l) 5. Does the hdmi audio device actually work? ;) PS. If you do try running, don't forget to correct the dtb entry in /boot/extlinux/extlinux.conf before you run it. The dtb file has definitely changed 3.13.x - 3.14.0.rcx). I don't habitually use HDMI audio, but I certainly can test it. Up to now with my self-built kernels and the ones from your repo, I've not been able to get it working so it will be interesting to try your newest kernel. I can probably do this tonight but, if not, then Saturday morning (US Pacific time). -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: The imx fb is no longer a stand-alone module and looks like it wont be created/initialized without a hdmi device being plugged in, which is a PITA from a jivelite running as a service, point of view. Also, I don't believe the hdmi audio device will be created without a hdmi device being detected. (Looks like there is a bunch of initialisation code, that parses ELD for supported sample rates, from the actual plugged-in device, but I've not really looked too closely.) Questions are With a hdmi device being plugged in from boot. 1. Video still works? 2. Does it work any better than it did before, vis-a-vis EDID detection code. (ISTR you had to set params on kernel cmd line.) 3. Does /dev/fb0 get created? 4. Do you get a hdmi audio device listed? (aplay -l) 5. Does the hdmi audio device actually work? ;) PS. If you do try running, don't forget to correct the dtb entry in /boot/extlinux/extlinux.conf before you run it. The dtb file has definitely changed 3.13.x - 3.14.0.rcx). I used a Wandboard QUAD for this experiment, which prior to the update was working with HDMI on 3.13.5-200.1.cs.fc20.armv7hl. The answer to all 5 questions is no. I left my mode setting kernel command line as it was. Had the video worked at all, I would have removed it, but it didn't. I am back on 3.13.5-200.1.cs.fc20.armv7hl now. If there is something else I should try, please let me know. Best regards, Charles. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] LMS on Cubox Arm7?
hagensieker wrote: Yes. Chose Arch to challenge myself a bit. This platform is a Cubox -4 I do have Ubuntu installed on a separate micro SD card and Squeezeboxserver works just by installing the .deb package. Bit tougher in arch. The armv6hf Squeezelite works perfectly and I do have another Squeezeboxserver so I'm by no means out of business but I sure would like to install the server on this Cubox so I can travel with it. Full desktop computer with an extremely small form factor. I do want to try a build of this for Arch if for no other reason than to learn from it. The best, most comprehensive, instructions are the ones found here: http://forums.slimdevices.com/showthread.php?99648-Howto-build-a-self-contained-LMS You still have the choice of building perl 5.14.2 from sources and using a nightly tar ball with pre-compiled CPAN libraries or keep using perl 5.18.2 but then build your *own* CPAN libraries. Fortunately, the instructions referenced above cover both procedures in reasonable detail. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=100980 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Hammer wrote: Hi, sometimes when I reboot the wandboard or if I lose power and the wandboard restarts after power is restored, my USB output is changed to mute. And I need to log on to the wandboard and run alsamixer -c3 to unmute the output. Is there a way to permanently disable mute and get volume to always be at 100 do I don't need to log on to the wandboard each time I reboot or lose power? Thank you! Hammer Code: alsactl store 3 This should be what you need to store the ALSA mixer settings for card #3. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] LMS on Cubox Arm7?
hagensieker wrote: I have Perl 5.18.2-2 Will the build work on this version and to be honest I've never done anything like this before. Are there any clear cut instructions for this? Sorry that my suggestion didn't work. I thought that support for 5.18 had been added, and it has, just not *binary* support for 5.18 in the daily builds. You have two choices, locally build a supported version of Perl (e.g., 5.14) or hack the buildme.sh script to try to build versions of the missing Perl modules that would work under 5.18. http://forums.slimdevices.com/showthread.php?99648-Howto-build-a-self-contained-LMS shows some details about getting away from your Platform's Perl and that might be a way to go. You have definitely chosen the path less traveled by going for Arch Linux. Had you gone for Fedora 19 or 20, you could have used the repositories maintained by Clive Messer from here http://www.communitysqueeze.org/ Then you could to yum install squeezeboxserver and be done. Alternatively, had you installed say Debian wheezy 7.4 which also supports armhf, you would have had a 5.14 system Perl and then you could have used the logitech nightly tar file per my original suggestion. But okay, you did choose ArchLinux to challenge yourself, right? -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=100980 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] LMS on Cubox Arm7?
hagensieker wrote: I just got a Cubox -i4 Pro and installed Arch Linux on it. Is it possible to get LMS installed on it. Perl version is 5.18 I did install squeezelite and pointed it at another server in the home and it works great. Also I have another SD card with Ubuntu 12.04 LTS which of course there is a .deb package for so I can run LMS just by card swapping if I need to. However, I'd like to install it on Arch arm7 if that is humanly possibly. What say the experts here? Why don't you try the ARM Linux Tarball (for Perl 5.8-5.14, ARM EABI) from the nightlies from here: http://downloads.slimdevices.com/nightly/?ver=7.8? Although this will not have been compiled with hard floating point, it should still work. As an ordinary user (not root), you could try something like the following: Code: wget http://downloads.slimdevices.com/nightly/7.8/sc/cedf697/logitechmediaserver-7.8.0-1391769631-arm-linux.tgz tar xvf ./logitechmediaserver-7.8.0-1391769631-arm-linux.tgz cd ./logitechmediaserver-7.8.0-1391769631-arm-linux ./slimserver.pl (The version number embedded in the name of the tar archive changes every night, so please adjust accordingly.) If this starts up LMS okay, you should install it in a more official place and create a startup script for it. If not, please report back here and someone else may have a better idea. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=100980 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
sideways2 wrote: BTW...can someone reply with the correct syntax for filling in the blanks on getting a shared Windows folder to be mounted?? I think I'm missing something as I get a red error mount.cifs: bad UNC Again...thanks...much appreciated!! Dave //HOSTNAME.local/ShareName or //DottedDecimal_IP/Sharename Ideally, you should be able to ping whatever lies between // and / and get a response (firewalls permitting). The most common mistake is to think that because we are talking to a Windows share we need backslashes; forward slashes are, in fact, mandatory. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
Triode, I've done a bit more to tidy up the vertical positioning of the UI elements in the different scaling of HDSkin. I was aiming to acheive the following: 1) Get all UI elements on the right half of the screen to stay within the vertical boundaries defined by the bottom and top edges of the album artwork (even when using non-widescreen aspect ratios). 2) Keep the metadata text confined to the top half of the vertical boundaries of the album artwork, hence each line of text is spaced apart by coverSize/6. 3) Spectrum analyzer space occupied should exactly match the space occupied by the album art, so that the transition between the two versions only differs in the content of that square. 4) Use a background texture to give the user an idea of the area occupied by the spectrum analyzer, even when it's not being driven hard by a signal. My current attempt at this could use some improvement, but I hope you get the idea. The four images attached are at two screen shots of 1280x720 (one with album art and one with visulization) and two screen shots of 1280x1024 (one with album art and one with visulization). The patch attached is against the very latest git version. +---+ |Filename: vertscale.patch.zip | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=15597| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: SQUEEZE-WEB-GUI-LUA-0.1-0.9.GITD107C4B Adrian has done some more work on the lua version of the web-gui Changes... Code: * Mon Feb 03 2014 - 0.9-d107c4b - Git branch master revision d107c4b2fae0f87836230e131c877f4fd64eacc2. handle relative symlinks for /etc/localtime, fix widths of selection box. * Mon Feb 03 2014 - 0.8-81b0436 - Git branch master revision 81b0436df1daa9fccf146a06b22a366ab881cb60. Change copyright, 2014 - 2013-2014. (web design dates back to 2013). - Git branch master revision 7b4b8f3662519dc9083e8de7bdd72e6243aaa522. Add image width and height tags to gpl image. - Git branch master revision ed867098e86d9f730227fd42bcd28e09c6685d19. Put gpl image and link back in Footer. * Mon Feb 03 2014 - 0.7-6fcb78e - 'Requires: squeeze-player-scripts = 1-2' for sp-localConfig and sp-hostnameConfig. - Git branch master revision 6fcb78e56ace728732a21a18f55a1ed2966cb413. update system storage and menu. wrap captures in coroutines to help prevent blocking. reorder squeezelite menu to hide more complex options. - Switch back to git master branch. Update... Code: sudo csos-cleanUpdate-testing squeeze-web-gui-lua This is really shaping up very nicely! I really like the hidden advanced menu options for squeezelite, and I have been able to successfully mount NFS and CIFS shares with no trouble at all (the red error feedback text helps a lot). Of course, the snappy speed is very welcome, even if not stricly necessary! -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
@Triode, After Michael's changes to the 1920x1080 resolution skins, I wanted to offer the following patch that attempts to keep things on the Now Playing screen reasonably proportioned at lower resolutions too. I am offering a patch for HDSkinApplet.lua which, IMO, does a reasonable job at this. It even makes the 640x480 skin not completely unreasonable in now playing mode. Before and After screenshots are attached. However, I think the more useful improvement is to 1280x720, which now more closely matches the proportions of the 1920x1080 skin, which I though Michael had optimized quite nicely. The intent was to do no harm to the higher resolutions, so those should't have changed noticeably. +---+ |Filename: Screen Shot 2014-02-08 at 5.19.37 PM.jpg | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=15587| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
sideways2 wrote: Evening!! I'm trying to configure my first wandboard...having a little trouble with wireless... I can enter my router and pass with a save but a save and reboot doesn't come back up in order to power down as said in the instructions and I cannot see it in a network scan afterwards... But if I cycle power again with a wired connection and look under the wireless tab...my router\pass info is saved... Then...doing a halt and cycling power without a wired connection leaves me in limbo again... Wired works fine and it comes up fine on my Droid SB app... Dave Your SSID for the network should be available to select from the pull-down menu. When you say you entered your router, do you mean you typed it in or was it available for selection from the drop-down list next to Network Name? If you have a hidden network, you might have to type it in, otherwise I hope you found it available as one of the options in the drop-down selection box. Please confirm. Can you also confirm that you are _not_ using WEP as the wireless security encryption method on your network? I ask, because that is one that won't work on the Wandboard. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
I know shairport isn't part of this project, but I've spent two evenings on this and have to admit I'm not going find the answer without help. (1) testing from the command line Code: [root@wandquad ~]# cat /etc/sysconfig/shairport # OPTIONS=-a Shairport -o ao -- -d alsa -o dev=hw:0 OPTIONS=-p 9002 -a Shairport -o ao -- -d alsa -o dev=default:CARD=imxspdif # OPTIONS=-a Shairport -o dummy [root@wandquad ~]# . /etc/sysconfig/shairport [root@wandquad ~]# sudo -u shairport /usr/bin/shairport $OPTIONS Listening for connections. All is well, I can play music and avahi-browse -a -r shows the following: Code: = eth0 IPv4 2029E6B0BD56@ShairportAirTunes Remote Audio local hostname = [wandquad.local] address = [10.0.1.204] port = [9002] txt = [pw=false txtvers=1 vn=3 sr=44100 ss=16 ch=2 cn=0,1 et=0,1 ek=1 sm=false tp=UDP] (2) Using systemd Code: [root@wandquad ~]# unset OPTIONS [root@wandquad ~]# systemctl enable shairport ln -s '/usr/lib/systemd/system/shairport.service' '/etc/systemd/system/multi-user.target.wants/shairport.service' [root@wandquad ~]# systemctl start shairport [root@wandquad ~]# systemctl status -l shairport shairport.service - Shairport AirTunes receiver Loaded: loaded (/usr/lib/systemd/system/shairport.service; enabled) Active: active (running) since Tue 2014-02-04 07:23:28 PST; 5s ago Main PID: 2948 (shairport) CGroup: /system.slice/shairport.service #9492;#9472;2948 /usr/bin/shairport -p 9002 -a Shairport -o ao -- -d alsa -o dev default:CARD=imxspdif Feb 04 07:23:28 wandquad systemd[1]: Starting Shairport AirTunes receiver... Feb 04 07:23:28 wandquad systemd[1]: Started Shairport AirTunes receiver. Looks okay, but the service is continually failing and restarting Code: [root@wandquad ~]# tail -10 /var/log/messages Feb 4 07:23:27 wandquad shairport: FATAL: MDNS child process died unexpectedly! Feb 4 07:23:27 wandquad shairport: Listening for connections. Feb 4 07:23:27 wandquad shairport: Shutting down... Feb 4 07:23:27 wandquad systemd: shairport.service: main process exited, code=exited, status=1/FAILURE Feb 4 07:23:27 wandquad systemd: Unit shairport.service entered failed state. Feb 4 07:23:28 wandquad systemd: shairport.service holdoff time over, scheduling restart. Feb 4 07:23:28 wandquad systemd: Stopping Shairport AirTunes receiver... Feb 4 07:23:28 wandquad systemd: Starting Shairport AirTunes receiver... Feb 4 07:23:28 wandquad systemd: Started Shairport AirTunes receiver. Feb 4 07:23:53 wandquad shairport: WARNING: couldn't create avahi client: Daemon not running! The claim that the avahi-daemon is not running doesn't seem true, but shairport is failing under systemd due to not being able to advertise itself using MDNS. Code: [root@wandquad ~]# systemctl status avahi-daemon.service avahi-daemon.service - Avahi mDNS/DNS-SD Stack Loaded: loaded (/usr/lib/systemd/system/avahi-daemon.service; enabled) Active: active (running) since Thu 1970-01-01 00:00:10 PST; 44 years 1 months ago Main PID: 385 (avahi-daemon) Status: Server startup complete. Host name is wandquad.local. Local service cookie is 348406120. CGroup: /system.slice/avahi-daemon.service #9500;#9472;385 avahi-daemon: running [wandquad.local] #9492;#9472;389 avahi-daemon: chroot helper I am mystified as to why running from the command line as user shairport should be any different from running via systemd (also as user shairport). I can get the shairport to stay running under systemd by telling it to use a different MDNS backend (--mdns=tinysvcmdns), but although it advertises itself, I can't get music to play. I really should be able to use the installed and running avahi-daemon. Any help would be appreciated. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Rewind! ;) F19 or F20? Using the shairport-1.0-0.2.dev.20140125git4f5af6f rpm or something you have done on your own? EDIT: And something that comes to mind from the last time there was any shairport discussion, you have stopped squeezeboxserver service before you run shairport service? F20 (see title of my post). Your shairport rpm (not self-built). Yes, I stopped squeezeboxserver for these experiments, but I am long-term planning to use port 9002 (-p 9002) for shairport which should avoid a clash (considering I also comment out the Conflicts= line in the service definition for shairport). -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Not tried to play anything to it yet.. Code: [fedora@wandboard system]$ sudo systemctl status shairport -l shairport.service - Shairport AirTunes receiver Loaded: loaded (/usr/lib/systemd/system/shairport.service; enabled) Active: active (running) since Tue 2014-02-04 20:50:22 GMT; 1min 45s ago Main PID: 1126 (shairport) CGroup: /system.slice/shairport.service #9500;#9472;1126 /usr/bin/shairport -a ShairportWAND -p 9002 -o ao -- -d alsa -o dev hw:2 #9492;#9472;1127 avahi-publish-service 2ADC8B457798@ShairportWAND _raop._tcp 9002 tp=UDP sm=false ek=1 et=0,1 cn=0,1 ch=2 ss=16 sr=44100 vn=3 txtvers=1 pw=false Feb 04 20:50:22 wandboard systemd[1]: Started Shairport AirTunes receiver. Feb 04 20:50:23 wandboard shairport[1126]: Established under name '2ADC8B457798@ShairportWAND' $ ps axfwww 1126 ?Ss0:00 /usr/bin/shairport -a ShairportWAND -p 9002 -o ao -- -d alsa -o dev hw:2 1127 ?S 0:00 \_ avahi-publish-service 2ADC8B457798@ShairportWAND _raop._tcp 9002 tp=UDP sm=false ek=1 et=0,1 cn=0,1 ch=2 ss=16 sr=44100 vn=3 txtvers=1 pw=false $ cat /etc/sysconfig/shairport OPTIONS=-a ShairportWAND -p 9002 -o ao -- -d alsa -o dev=hw:2 $ avahi-browse -a -r = wlp2s0 IPv4 2ADC8B457798@ShairportWANDAirTunes Remote Audio local hostname = [wandboard.local] address = [192.168.0.56] port = [9002] txt = [pw=false txtvers=1 vn=3 sr=44100 ss=16 ch=2 cn=0,1 et=0,1 ek=1 sm=false tp=UDP] Long story short, all my troubles were down to selinux policy being left at the Fedora default of enforcing. I now have it set to permissive. I really appreciate you looking into this. Your success gave me the confidence that it *should* work, and eventually I was able to track down the issue. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Code: sudo yum install glibc-devel Yes, working now. Thanks. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: dsdreamer, great that you have hdmi working with F20 and that 3.13 kernel. If you can, (because I haven't tested it), and chance of you testing spdif TOSLINK out and verifying that is working? Also, perhaps you could install cpuburn package and run cpuburn-a9 for 24 hours for a little torture testing. I have tested Toslink and it is fine for source material up to 96kHz. Higher sample rates ( 96kHz) downsample to 44.1kHz, or 48kHz, rather than the 88.2kHz or 96kHz that I assume would be possible and preferable. This is without specifying any resampling strategy to Squeezelite. I also briefly tested NFS mounting via the new page on squeeze-web-gui-lua. It works fine for making a new NFS mount providing the mount point /storage was already made. Right now error messages from failed attempts are not fed back to the user; one needs to go looking the the logs. Also, unmount works fine from the GUI. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
dsdreamer wrote: I have tested Toslink and it is fine for source material up to 96kHz. Higher sample rates ( 96kHz) downsample to 44.1kHz, or 48kHz, rather than the 88.2kHz or 96kHz that I assume would be possible and preferable. This is without specifying any resampling strategy to Squeezelite. I tried to reproduce playing a 96kHz track via Toslink, and it plays at 48kHz. I have to assume that I was mistaken in thinking that I read 96kHz on the DAC's display last night. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
dsdreamer wrote: This is way down in terms of importance, but I noticed that the Jivelite screen that reports the asound stream0 status info doesn't anymore (for my particular DAC). We had similar issues before, and at that time we found that the exact format of /proc/asound/card[n]/stream0 varies with kernel, and the assumptions in the Lua code that parses this file may have to be adjusted for different kernels. I am not sure if this is the same issue again. Code: TEAC Corporation TEAC UD-501 at usb-ci_hdrc.1-1.1, high speed : USB Audio Playback: Status: Running Interface = 1 Altset = 1 Packet Size = 63 Momentary freq = 44102 Hz (0x5.8340) Feedback Format = 16.16 Interface 1 Altset 1 Format: S32_LE Channels: 2 Endpoint: 1 OUT (ASYNC) Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000 Data packet interval: 125 us Interface 1 Altset 2 Format: S32_LE Channels: 2 Endpoint: 1 OUT (ASYNC) Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000 Data packet interval: 125 us Code: diff --git a/share/jive/applets/SetupSqueezelite/SetupSqueezeliteApplet.lua b/sh index c051f6e..4038332 100644 --- a/share/jive/applets/SetupSqueezelite/SetupSqueezeliteApplet.lua +++ b/share/jive/applets/SetupSqueezelite/SetupSqueezeliteApplet.lua @@ -310,8 +310,9 @@ function _parseStreamInfo(self, card) if t.status == Running then t.interface = parse(Interface = (%d+)) t.altset= parse(Altset = (%d+)) - skip(2) - t.momfreq = parse(Momentary freq = (%d+) Hz) + while not eof() and not t.momfreq do + t.momfreq = parse(Momentary freq = (%d+) Hz) + end t.feedbkfmt = parse(Feedback Format = (.*), true) end I could also have simply replaced skip(2) with skip(1) to just fix it on this particular kernel, but it may be better to get rid of all the magic integers used as arguments to skip? +---+ |Filename: jivelite.patch.zip | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=1| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Code: [root@wandquad squeeze-web-gui-lua]# luajit squeeze-web-gui.lua luajit: /usr/share/luajit-2.0.2/turbo/util.lua:153: librt.so: cannot open shared object file: No such file or directory stack traceback: [C]: in function 'load' /usr/share/luajit-2.0.2/turbo/util.lua:153: in main chunk [C]: in function 'require' /usr/share/luajit-2.0.2/turbo/log.lua:34: in main chunk [C]: in function 'require' /usr/share/luajit-2.0.2/turbo.lua:78: in main chunk [C]: in function 'require' squeeze-web-gui.lua:23: in main chunk [C]: at 0xc97c [root@wandquad squeeze-web-gui-lua]# yum downgrade turbolua Loaded plugins: langpacks, refresh-packagekit Resolving Dependencies -- Running transaction check --- Package turbolua.armv7hnl 0:1.0.0-1.fc20 will be a downgrade --- Package turbolua.armv7hnl 0:1.1.0-0.1.git8948f21.fc20 will be erased After downgrading, it now it works fine. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
This is way down in terms of importance, but I noticed that the Jivelite screen that reports the asound stream0 status info doesn't anymore (for my particular DAC). We had similar issues before, and at that time we found that the exact format of /proc/asound/card[n]/stream0 varies with kernel, and the assumptions in the Lua code that parses this file may have to be adjusted for different kernels. I am not sure if this is the same issue again. Code: TEAC Corporation TEAC UD-501 at usb-ci_hdrc.1-1.1, high speed : USB Audio Playback: Status: Running Interface = 1 Altset = 1 Packet Size = 63 Momentary freq = 44102 Hz (0x5.8340) Feedback Format = 16.16 Interface 1 Altset 1 Format: S32_LE Channels: 2 Endpoint: 1 OUT (ASYNC) Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000 Data packet interval: 125 us Interface 1 Altset 2 Format: S32_LE Channels: 2 Endpoint: 1 OUT (ASYNC) Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000 Data packet interval: 125 us -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: One thought did occur to me. You know you need the fb device to even get jivelite to launch on the tty, and this is not automatically modprobe'd? For months now, I've been doing it out of rc.local. No, I was not aware, but I had been trying to modprobe likely-sounding modules such as imx_hdmi without making any headway. Using your wandboard.local rpm, and 3.13.0-2.4.cs.fc20.armv7hl, I have my display back again! Actually, I also needed to make a change in /boot/extlinux/extlinux.conf: Code: append video=HDMI-A-1:1920x1080-24@60 to get native resolution out of the monitor. I also installed the other packages you recommeded. Thanks for setting me straight. I have now given Jivelite something more like the originally-envisaged workout, and everying looks fine, including visualizations. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Thank you. If you have time, might be helpful if you could give the jivelite F20 version a good workout. Okay, I can't claim a good workout, but it looks good so far. I will do more thorough testing this evening after work. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Yes, that's expected. Thank you. If you have time, might be helpful if you could give the jivelite F20 version a good workout. Have gone back to using system packages for the lua module dependencies on F20 having made lua51 installable alongside the default 5.2. socket, filesystem, expat, loop, md5, zipfilter, cjson, several being later versions than those supplied with jivelite, that we are still using for the F19 build. It would be good to be able to say to Triode with some confidence that the later versions don't cause any issues, (that we didn't have already), and ask him to upgrade the modules supplied with the jivelite source distribution. (F19 build is still using the private module versions, supplied by jivelite and installed to /usr/share/jive, rather than system versions installed to /usr/share/lua/5.1 and /usr/lib/lua/5.1.) Running X-protocol over ssh, I have checked it out on some of the smaller skins. I would do more testing, but I am not able to get HDMI out to work with 3.13.0-2.2.cs.fc20.armv7hl, using the standard Fedora boot scripts, the only hdmi-related lines from dmesg are the following. Code: [8.314872] imx_hdmi: module is from the staging directory, the quality is unknown, you have been warned. [8.316473] imx-hdmi 12.hdmi: Detected HDMI controller 0x13:0xa:0xa0:0xc1 However, this is likely down to my own incompetence in setting up the F20 image, rather than anything you need to deal with. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Code: sudo systemctl restart squeeze-web-gui-lua Then access via http://wandboard.local:8081;. Not worried about functionality testing. Don't go there. ;) Only that the web page appears. (Of course, if not using bonjour/mdns/avahi, http://ip_address:8081;.) Yes, this web page comes up and looks fine, and shows Copyright © 2014 Adrian Smith (code) and Clive Messer (web design) JackOfAll wrote: SQUEEZE-WEB-GUI-1-69.20140129GITE3254B5 squeeze-web-gui is the original Java Web-GUI, which used to be packaged as community-squeeze-web. Code: sudo csos-cleanInstall-testing squeeze-web-gui The path used to be http://wandboard.local:8080/CommunitySqueeze/;, now shortened to http://wandboard.local:8080/squeeze/;. If not accessing via port 8080 (tomcat directly), you'll need to reload the apache proxy config. (It will be added to the postinstall scriplet, the next time I make a build.) Code: sudo systemctl reload httpd This also looks fine. Claims to be version 1-68 (-c.f.- 1-69 from the package name). . Tested using the following packages Code: jivelite.armv7hnl 1:0.1-0.72.20140123git427e8a0.fc20 @community-squeeze squeeze-web-gui.noarch 1-69.20140129gite3254b5.fc20 @community-squeeze squeeze-web-gui-lua.noarch 0.1-0.2.gitad9f8e2.fc20 @community-squeeze No issues found. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: Squeezelite - a small headless squeezeplay emulator for linux (alsa only)
JackOfAll wrote: Squeezelite already does DSD-PCM conversion and can output DoP if given the -D flag. I beleive the original poster is asking for on-the-fly PCM-(DSD in a PCM container, e.g., DoP) conversion for rendering by a DSD-capable DAC. This could be summarized as PCM-DSD-DoP. Some people like the euphonic smoothness of DSD-rendered material, even when there is no souce-content information above 20kHz. Before you all pick up bricks to throw at me, I am not making any claims that this is a good and useful idea, just that some people might want to experiment with it. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=97046 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Alcaudon wrote: I'm a total noob to linux (this is the first time I use it), and all the knowledge I have comes from a few quick and dirty google searches done in 30 minutes but probably I can help you. [snip] Then I just did: Code: [fedora@wandboard ~]$ sudo mount NASIPaddress:/volume/folder This worked for me (the music folder in my NAS was listed in the /storage folder), but as I said I have no clue of linux so all this might be totally wrong, but it could be of help until someone else with more knowledge than I gives you a better reply. Good luck! ;) That's fine and thanks for helping out a forum member. Having edited fstab, you could reduce typing a bit by just doing: Code: sudo mount -a This will mount anything in fstab that is not already mounted. Thus, if you added a new line in /etc/fstab (or uncommented one), it will incrementally execute just that line. To verify, use Code: df -t nfs The newly mounted nfs filesystem and its usage stats will be displayed assuming the mount succeeded. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
slackhead wrote: Jumping in lists works pretty well - the sensitivity problem you mention is very apparent, but I have a feeling this maybe a FLIRC problem as I've had the same thing when using FLIRC on other devices/application. Perhaps this is something we could take up with the FLIRC developers? A while back, I found things improved greatly by setting the Inter-key delay parameter to 3 (default is 6). Some people have reported a value of 4 works, and I tried that first which was an improvement, but 3 seems to be optimum for me. I have attached a version of ralphy's keymap unchanged except for the inter-key delay being set to 3. If you have loaded this keymap, you could verify the change by running the command line and listing the settings as shown below. Code: sudo ./flirc_util settings Settings: sleep detection: Disabled noise canceler: Disabled inter-key delay: 3 state:0 builtin profiles: NA I'd be interested to know if other Flric users with the SB remote see this as a useful change. +---+ |Filename: ralphy_jivelite_flirc_map_debounce.fcfg.zip | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=15505| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
slackhead wrote: Thanks, but it's not helping here - I've tried changing that setting in the past to no avail. I'll get on the FLIRC forum sometime and ask if there is anything else that could help. Understood. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Triode wrote: I've added some code to emulate the old ir behaviour using the keyboard (and hence flirc) - would be good if Clive could add to the testing jivelite so people can tell me if they like it and if they find any side effects... Good idea, this works for me (tested in HD grid skin). I used it to search for short strings and was quite happy with its behavior in this limited use-case. Longer search strings are not going to be worth the pain, but I can live that since 4 or 5 letters is typically sufficient to narrow down the choices to a short scrollable list. I also tried jumping to places in the alphabet when browsing album artwork in grid mode and the list of artists. What fools me sometimes is when intital letters are missing from my collection (such as nothing with the intial letter Q), so I expect to 777 to get to R, but this overshoots; I have to know -a priori- that Q is missing and press 77 to get R instead. Similarly, I only need to press 99 (not 999) to get to Y. Perhaps if the whole thing slowed down, the user would understand that he or she had jumped over a missing letter on his way to R or Y. Alternatively, you could defer any jump-to-letter action until a multiple-key-press timer had expired, which is the way texting used to work in the dark ages. Finally, I would rather the letter overlay pop-up in the middle of the screen lingered a bit longer after I have selected a letter by numeric keypad presses. It often disappears before I've really seen it. It would be ideal if the timer that keeps the letter displayed be interuptable by the next keypress, but otherwise stays up long enough for the user to read it. Alternatively a group of keypresses should be resolved into a letter (after a timeout exprires) before any jumping to a new location occurs, and that one letter should be displayed for a longer time. Just my thoughts as I played with the new functionality. Others may feed differently. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: but clear (delete contents) of alsa params field (::16:), Save and Conditionally Restart. JackOfAll, Any chance we could, in a future release, drop this cunning default to ::16: ? It it trips up more people than it helps, IMO. I actually find on my Wandboards that the levels of distortion to the analog jack sound about the same with or without this workaround, anyway. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: dsdreamer, did you have a play with Ralphy's enhancements to your FLIRC config. Still wondering whether to make that a default config from the website jivelite page and drop the links to Triode's (which yours was based on???) and yours? Thoughts? I am willing to make a comparison and report back here. IMO, we need two FLIRC configuration files: one configuration tuned for HD Grid skin and one for all other list-based skins. Best regards, -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Whatever you (and others) think best. I don't use FLIRC. Of the three posted keymaps, Ralphy's seems the most comprehensive. Code: __ralphy.keys: 654A9E6B ` --- dsdreamer.keys: 654A433D = ___ralphy.keys: 654A433D = ___triode.keys: 654A433D = --- dsdreamer.keys: 6506F08A - ___ralphy.keys: 6506F08A - ___triode.keys: 6506F08A - --- dsdreamer.keys: 65167515 , ___ralphy.keys: 65167515 , ___triode.keys: 65167515 , --- ___ralphy.keys: 65D22341 ; --- dsdreamer.keys: 65169435 : ___ralphy.keys: 65169435 : --- ___ralphy.keys: 658E6FA6 : --- dsdreamer.keys: 65D2C049 . ___ralphy.keys: 65D2C049 . ___triode.keys: 65D2C049 . --- ___ralphy.keys: 65167306 [ --- dsdreamer.keys: 65D2E178 [ ___ralphy.keys: 65D2E178 [ --- ___ralphy.keys: 65D22163 } --- dsdreamer.keys: 65D2C258 0 ___ralphy.keys: 65D2C258 0 ___triode.keys: 65D2C258 0 --- dsdreamer.keys: 651677F3 1 ___ralphy.keys: 651677F3 1 ___triode.keys: 651677F3 1 --- dsdreamer.keys: 654A3F50 2 ___ralphy.keys: 654A3F50 2 ___triode.keys: 654A3F50 2 --- dsdreamer.keys: 658E9202 3 ___ralphy.keys: 658E9202 3 ___triode.keys: 658E9202 3 --- dsdreamer.keys: 658EF10C 4 ___ralphy.keys: 658EF10C 4 ___triode.keys: 658EF10C 4 --- dsdreamer.keys: 65D244BF 5 ___ralphy.keys: 65D244BF 5 ___triode.keys: 65D244BF 5 --- dsdreamer.keys: 658E90F3 6 ___ralphy.keys: 658E90F3 6 ___triode.keys: 658E90F3 6 --- dsdreamer.keys: 65D2E3A5 7 ___ralphy.keys: 65D2E3A5 7 ___triode.keys: 65D2E3A5 7 --- dsdreamer.keys: 65D242B0 8 ___ralphy.keys: 65D242B0 8 ___triode.keys: 65D242B0 8 --- dsdreamer.keys: 65169662 9 ___ralphy.keys: 65169662 9 ___triode.keys: 65169662 9 --- dsdreamer.keys: 658EF3EA a ___ralphy.keys: 658EF3EA a ___triode.keys: 658EF3EA a --- dsdreamer.keys: 658E94E0 B ___ralphy.keys: 658E94E0 B ___triode.keys: 658E94E0 b --- dsdreamer.keys: 654A412E c ___ralphy.keys: 654A412E c ___triode.keys: 654A412E c --- ___ralphy.keys: 655AC6B8 C --- dsdreamer.keys: 65D2C436 down ___ralphy.keys: 65D2C436 down ___triode.keys: 65D2C436 down --- ___ralphy.keys: 651614FC escape --- dsdreamer.keys: 658E8EC6 h ___ralphy.keys: 658E8EC6 h ___triode.keys: 658E8EC6 h --- dsdreamer.keys: 658E7393 left ___ralphy.keys: 658E7393 left ___triode.keys: 658E7393 left --- dsdreamer.keys: 654A20E1 p ___ralphy.keys: 654A20E1 p ___triode.keys: 654A20E1 p --- ___ralphy.keys: 654AA247 Q ___triode.keys: 654AA247 Q --- dsdreamer.keys: 65D22550 right ___ralphy.keys: 65D22550 right ___triode.keys: 65D22550 right --- dsdreamer.keys: 65D2469D up ___ralphy.keys: 65D2469D up ___triode.keys: 65D2469D up --- dsdreamer.keys: 658EF5F9 Z ___ralphy.keys: 658EF5F9 Z ___triode.keys: 658EF5F9 z Raphy seems to have a superset of IR codes compared to buttons on the SBT remote, of which there are only 29. Ralphy has 34 keys mapped, some of which I presume belong to older SB remotes. I had 26 keys mapped and Triode had 25. Given this information, I don't see the added value of keeping the less complete sets of key-bindings, from either myself or Triode, but I won't presume to speak for Triode. Just my suggestion. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
slackhead wrote: I might be missing something :-| but when using Jivelite on a 'grid' display mode with a FLIRC - how do you go into a menu item? Seems an 'enter' key is needed, but none of the posted configs have that so I've just changed the power button to do 'enter', but doesn't seem ideal (or intuitive for the better half ). Also got caught out by that damn Quit button again, which closes Jivelite and stops the web gui (including SBS) and ssh from working, but music keeps playing. Need to pull power. And (there's more), I've been trying the Visulizer (!) option, but like Albertone74 I can't see it (in any skin). Slackhead, I got around this by mapping the remote's play key to keyboard enter. However, you still need a back or escape key, and now I really have run out of reasonable keys to use, so I mapped the remote's sleep key to keyboard escape. Given that the pause button is a toggle between play and paused states, this is usable for me. For me, the quit button always closed Jivelite, but caused it to immediately respawn due to the jivelite user's autologin arrangement. I haven't seen the issue you report of getting into a state where you effectively lose network connectivity to the Wandboard. I can confirm that the visualizer option is MIA on the new R3 image, even though squeezelite is able to open the shared memory, Jivelite doesn't seem to find it. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Broken for me as well. Sorry, I'm entertaining at the moment so can't look into this. Quick look at /var/log/messages Code: Jan 19 16:49:57 wandboard jivelite: WARN jivelite.vis - SpectrumMeter.lua:116 ** 1: 21 2: 21 Who's responsible for this cryptic dung? :) It's important right, that's why it's being logged as a warning. Warning me of what? ** 1: 21 2: 21... WTF does that mean? Oh right, it's a message warning me that I need to go look at line 116 of SpectrumMeter.lua to even have half a clue what it's trying to tell me. Face palm! Observation: If I start Jivelite as user fedora, I can get visualizations to work, whereas if I do so under user jivelite, I can't. This was tested using ssh -XYC fedora@wandboard.local and starting jivelite from the command prompt. I tried adding jivelite to group wheel, but that didn't seem to help. I am puzzled, because the permissions for the shared memory look fine: Code: ls -l /dev/shm/ total 36 -rw-rw-rw- 1 squeezelite squeezelite 32820 Jan 19 16:54 squeezelite-00:1f:7b:b4:01:a5 -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackofAll, I've been running the F19 R3 image for a Wandboard Quad. Only tested on Ethernet with my TEAC USB DAC so far. I've tested LMS scanning my media library over NFS, and played back music at various sample rates, including DSD playback, both via DoP and transcoded to PCM. I've configured via the new cs-web GUI on port 8081 and via the old one on port 8080. No problems, except that I don't have access to hw:CARD= devices on the new cs-web, but have to use front:CARD= instead. I wasn't sure if these were functionally equivalent, although I checked that they both play music. JIvelite was tested via X using ssh -XYC jivelite@wandquad.local, and it works fine except you've forgotten to blow away the preference files, so the initial language setup and select skin dialogs don't happen. (guestfish --rw -a CSOS-Wandboard-Quad-F19-R3-20140117-1.img can be used to fix this on the raw image file without messing around with SD cards) So overall, a very smooth experience. Anything in particular I should check in addition to the above? -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
dsdreamer wrote: Eight bells and all's well. Code: [root@wandquad ~]# uname -a Linux wandquad 3.13.0-0.rc8.git1.2.1.cs.fc20.armv7hl #1 SMP Tue Jan 14 17:32:29 GMT 2014 armv7l armv7l armv7l GNU/Linux [root@wandquad ~]# uptime 07:30:52 up 8:08, 1 user, load average: 0.00, 0.01, 0.05 Make that 32 hours uptime now. I guess we could conclude this is stable. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Eight bells and all's well. Code: [root@wandquad ~]# uname -a Linux wandquad 3.13.0-0.rc8.git1.2.1.cs.fc20.armv7hl #1 SMP Tue Jan 14 17:32:29 GMT 2014 armv7l armv7l armv7l GNU/Linux [root@wandquad ~]# uptime 07:30:52 up 8:08, 1 user, load average: 0.00, 0.01, 0.05 -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: LOL. Over-inflated load-avg with 3.0.35, and under-inflated with 3.13. Another task for you if you don't mind. I'm just finishing twiddling with what will be a F19 R3 image. If I PM you a link in a while, can you run that for me? (The current plan is that when the image download is back on-line via the website, anticipating Friday now, there will be the latest and greatest, with an up to date F19 and the elusive F20 image.) Yes, very happy to do that, will be looking for your PM. Just keep in mind I don't get free time until evening US Pacific time. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Before switching back to F19, while you are still on F20. Code: sudo csos-cleanUpdate-testing squeezelite It will ask to install libdsd2pcm as a dependency, now a shared lib rather than compiled into squeezelite. If you could test DSD functionality, everything is still working, DSD - PCM and DSD - DoP, that would be good! Got it. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Before switching back to F19, while you are still on F20. Code: sudo csos-cleanUpdate-testing squeezelite It will ask to install libdsd2pcm as a dependency, now a shared lib rather than compiled into squeezelite. If you could test DSD functionality, everything is still working, DSD - PCM and DSD - DoP, that would be good! Code: rpm -q squeezelite squeezelite-1.5-2.fc20.armv7hnl DSD- PCM works, but outputs at 352.8 kHz. (Is that expected?). Sounds fine. DSD- DoP works. 44.1, 88.2, 96 and 192kHz FLAC material also tested and no issues found. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Triode, I accidentally downloaded a new album in ALAC format last night, that happens to be encoded in ALAC in 24/96kHz resolution. When I try to play it on Wandboard, I had trouble getting the first track to start, but I did eventually succeed but it failed to transition from the 1st to the second track. The log file from output=debug decode=debug when starting a track is shown below. Code: [07:16:45.176491] output_init_alsa:636 init output [07:16:45.177302] output_init_alsa:658 requested alsa_buffer: 40 alsa_period: 4 format: any mmap: 1 [07:16:45.177422] output_init_common:334 outputbuf size: 3528000 [07:16:45.184663] output_init_common:394 supported rates: 384000 352800 192000 176400 96000 88200 48000 44100 [07:16:45.277594] output_init_alsa:670 memory locked [07:16:45.278683] output_thread:465 open output device: hw:CARD=UD501 [07:16:45.278719] output_init_alsa:693 set output sched fifo rt: 45 [07:16:45.279914] alsa_open:234 opened device hw:CARD=UD501 using format: S32_LE sample rate: 44100 mmap: 1 [07:16:45.280144] alsa_open:313 buffer: 40 period: 4 - buffer size: 1764 period size: 441 [07:16:45.285634] output_vis_init:135 opened visulizer shared memory as /squeezelite-00:1f:7b:b4:01:a5 [root@wandquad squeezelite]# head -100 squeezelite.log [07:16:45.176491] output_init_alsa:636 init output [07:16:45.177302] output_init_alsa:658 requested alsa_buffer: 40 alsa_period: 4 format: any mmap: 1 [07:16:45.177422] output_init_common:334 outputbuf size: 3528000 [07:16:45.184663] output_init_common:394 supported rates: 384000 352800 192000 176400 96000 88200 48000 44100 [07:16:45.277594] output_init_alsa:670 memory locked [07:16:45.278683] output_thread:465 open output device: hw:CARD=UD501 [07:16:45.278719] output_init_alsa:693 set output sched fifo rt: 45 [07:16:45.279914] alsa_open:234 opened device hw:CARD=UD501 using format: S32_LE sample rate: 44100 mmap: 1 [07:16:45.280144] alsa_open:313 buffer: 40 period: 4 - buffer size: 1764 period size: 441 [07:16:45.285634] output_vis_init:135 opened visulizer shared memory as /squeezelite-xx:xx:xx:xx:xx:xx (obfuscated by dsdreamer) [07:16:45.285761] decode_init:129 init decode [07:16:45.285974] register_dsd:624 using dsd [07:16:45.294611] register_pcm:382 using pcm [07:16:45.384979] codec_open:211 codec open: 'l' [07:16:45.385254] ff_open_alac:566 open alac [07:16:45.901608] ffmpeg: stream 0, timescale not set [07:16:45.901906] ff_decode:288 format: name:mov,mp4,m4a,3gp,3g2,mj2 lname:QuickTime / MOV [07:16:45.935466] ffmpeg: stream 0, offset 0x8136e: partial file [07:16:45.935998] ff_decode:308 found stream: 0 [07:16:45.936330] ff_decode:340 setting track_start [07:16:45.936453] ffmpeg: Error : 506925 bits left [07:16:46.047794] _output_frames:59 start buffer frames: 24576 [07:16:46.048080] _output_frames:140 track start sample rate: 96000 replay_gain: 0 [07:16:46.058238] output_thread:465 open output device: hw:CARD=UD501 [07:16:46.066997] alsa_open:234 opened device hw:CARD=UD501 using format: S32_LE sample rate: 96000 mmap: 1 [07:16:46.067404] alsa_open:313 buffer: 40 period: 4 - buffer size: 3840 period size: 960 [07:16:46.099072] ff_decode:377 avcodec_decode_audio4 error: -1 Operation not permitted [07:16:46.099347] ffmpeg: stream 0, offset 0x8503a: partial file [07:16:46.099434] ff_decode:360 av_read_frame error: -1094995529 Invalid data found when processing input [07:16:46.099516] ffmpeg: stream 0, offset 0x88cbe: partial file [07:16:46.099636] ff_decode:360 av_read_frame error: -1094995529 Invalid data found when processing input [07:16:46.099717] ffmpeg: stream 0, offset 0x8c91d: partial file [07:16:46.099790] ff_decode:360 av_read_frame error: -1094995529 Invalid data found when processing input [07:16:46.099871] ffmpeg: stream 0, offset 0x90581: partial file [07:16:46.099943] ff_decode:360 av_read_frame error: -1094995529 Invalid data found when processing input [07:16:46.100015] ffmpeg: stream 0, offset 0x941e0: partial file [07:16:46.100086] ff_decode:360 av_read_frame error: -1094995529 Invalid data found when processing input The same files play fine on other media players. If needed I can send you a sample to assist with debugging. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: OK, before anyone starts chasing rainbows, as this might be down to a specific version of ffmpeg and I think that we already might be on different version of 2.x.x, depending on whether it is the ffmpeg2 private package I made for F19, or the Fedora system package where ffmpeg is now v2 on F20, and you might be running on F20 and not F19.. Code: rpm -q squeezelite ffmpeg ffmpeg2 I agree, please don't anyone work on this until get a chance to verify, which be my evening US Pacific time. I am definitely using F20, but I don't know which repo supplied by ffmpeg[2] dependencies. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: Maybe this helps you guys, maybe it doesn't. (But it might help me, if you guys report back that you run it with a stable uptime of xx days.) Although I still intend to release the CSOS F20 image with a 3.12.6 custom kernel, 3.13 is at rc8 stage, so I've just started re-basing my 200+ patch set with the expectation that we will see a 3.13 release, real soon now. The first 3.13 kernel I made is based on the Fedora rawhide 3.13rc8 kernel package. It has a basic patch set, but should give you working wi-fi, (subject to the caveat that you need to power cycle reboot, or use the button on the motherboard, after halt), SATA, HDMI video (but based on the original staging driver, the most recent needs work, to make it work), and SPDIF output. Code: sudo csos-cleanUpdate brcmfmac-sdio-firmware sudo csos-cleanUpdate-testing kernel Before you reboot, you'll need to fix-up the /boot/extlinux/extlinux.conf. A new entry will have been added for the new kernel, but it'll be pointing at the old dtb location. (There is still a Fedora bug here!) Although it will probably boot, pointing at the old device tree file, you'll need to be using the new device tree file to get the new functionality! (snip) I will be happy to test the kernel you built and report on the stability over the next few days. BTW, I did build my own kernel this last weekend using Robert C Nelson's instructions and patchset, and thus I've already had 3.13.rc7 working on my Wandboard Quad, which did allow me to enable WiFi. However, I reverted to the stock Fedora 3.12.6 kernel since I was really not planning to get into kernel building for any serious purpose. Probably, the main benefit is that I now have a serial cable to USB adapter that works :D -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: F20, it's either going to be the Fedora ffmpeg-2.1.1-1 or CS ffmpeg-2.1.1-1.1, where I have additionally enabled faac and fdk-aac, which aren't enabled in the Fedora build. Maybe you should try from the cmd line, ffmpeg -i /path/to/your/alac/alac.m4a output.wav, and see if that barfs on converting one of your new files, from alac to wav. Code: rpm -q ffmpeg-libs ffmpeg-libs-2.1.1-1.1.fc20.armv7hnl Also, Code: ffmpeg -i /mnt/music/flac/Miles\ Davis/01-01-Miles_Davis-Generique-SMR.m4a -f s32le - | aplay -f S32_LE -r 96000 -c 2 -D hw:CARD=UD501,0 plays the music very nicely with hesitation, deviation or repetitiion. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Triode wrote: Hum - don't know why it doesn't like that - can you try this change to the code: Code: --- plugins/BBCiPlayerRTMP/RTMP.pm 2013-07-29 21:01:50.0 +0100 +++ Plugins/BBCiPlayer/RTMP.pm 2014-01-12 11:42:45.0 + @@ -1426,7 +1426,11 @@ $start = $newtime; - $song-can('startOffset') ? $song-startOffset($newtime) : $song-{startOffset} = $newtime; + if($song-can('startOffset')) { + $song-startOffset($newtime); + } else { + $song-{startOffset} = $newtime; + } $client-master-remoteStreamStartTime(Time::HiRes::time() - $newtime); } I applied your patch at the specified place and then got a similar error message at line 107, so I applied the same fix there as well. So now, my overall patch looks as follows: Code: --- RTMP.pm.orig2014-01-12 09:56:15.874169043 -0800 +++ RTMP.pm 2014-01-12 09:58:15.571103959 -0800 @@ -104,7 +104,11 @@ $params-{'start'} = $newtime; - $song-can('startOffset') ? $song-startOffset($newtime) : $song-{startOffset} = $newtime; + if($song-can('startOffset')) { + $song-startOffset($newtime); + } else { + $song-{startOffset} = $newtime; + } $args-{'client'}-master-remoteStreamStartTime(Time::HiRes::time() - $newtime); } @@ -1426,7 +1430,11 @@ $start = $newtime; - $song-can('startOffset') ? $song-startOffset($newtime) : $song-{startOffset} = $newtime; + if($song-can('startOffset')) { + $song-startOffset($newtime); + } else { + $song-{startOffset} = $newtime; + } $client-master-remoteStreamStartTime(Time::HiRes::time() - $newtime); } As to why this change may be needed, I think the Ternary ?: is only allowed as an lvalue if the 2nd and 3rd arguments are guaranteed to be assignable, which is not true here. +---+ |Filename: mypatch.patch| |Download: http://forums.slimdevices.com/attachment.php?attachmentid=15472| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: F20 arm repo is now 100% in sync, with package versions in F19 arm repo, with the push of the latest lms-CPAN package to testing. The only thing missing is the custom kernel. Code: sudo systemctl disable firewalld.service sudo yum -y install http://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-20.noarch.rpm sudo yum -y install http://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-20.noarch.rpm sudo yum -y install http://www.communitysqueeze.org/repo/community-squeeze-repo.noarch.rpm sudo csos-cleanInstall-testing logitechmediaserver squeezelite jivelite Anyone non-technical, should be waiting, for the official image, rather than going with the upstream F20 minimal image plus the repo software, but feel free to suffer the same way as I have until you figure out where all the bodies are buried, and there are a few, issues that will have already been dealt with for you by the official CSOS image. But if you can live with a USB DAC and ethernet, have fun! I think you'll need to have a serial console to go through the initial setup, first time you boot the image, create a non-priv user, add to the wheel group, (so you can sudo). I don't know how you'll do that without serial console. Actually, I started with a minimal F20 image per the instructions here: https://fedoraproject.org/wiki/Architectures/ARM/F20/Installation#For_the_Wandboard_.28Freescale_i.MX6.29, also using the workable, but non-recommeded instructions to get SSH running for a first login without a serial cable, here: https://fedoraproject.org/wiki/Architectures/ARM/F20/Installation#FAQ. I actually did a general yum update before adding the rpmfusion repos and the CSOS repos, so I am therefore running Linux kernel 3.12.6-300.fc20.armv7hl. This allowed me to test whether I need the double alsa_open() workaround that Triode provided under the stock Fedora 3.12.6 Linux kernel. In fact, I don't, which I consider good news. I don't have any issues except that some network scripts used by the CSOS web-gui are not present on my system (/etc/sysconfig/network-scripts/ifcfg-wlan0 and /etc/sysconfig/network-scripts/ifcfg-eth0), which trips up the corresponding pages on the CSOS web-gui version 1-64. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Triode wrote: Thanks - plugin updated - could you test for me? Triode, thanks, that was quick! I just tested the particular use-case of jumping backwards and forwards in BBC iPlayer listen-again streams. This works fine now. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: I think I might have speculated at the time, that we were fixing something in the wrong place, that would go away anyway when moving to mainline kernel. ;) Yes, you certainly did. It will be interesting to know if lintweaker's Marantz SA-14S1 can also benefit from the kernel update and thus dispense with the need for additional user-space workarounds. I hope so. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Triode, I find I can can change the track position pointer when playing listen again streams, the pointer doesn't move, and the audio gets muted and the following error message appears: [14-01-11 22:29:52.1384] Slim::Control::Request::execute (1888) Error: While trying to run function coderef [Slim::Control::Commands::timeCommand]: [Can't modify non-lvalue subroutine call at /var/lib/squeezeboxserver/cache/InstalledPlugins/Plugins/BBCiPlayer/RTMP.pm line 1429.] This is on Logitech Media Server Version: 7.8.0 - 1388931871 @ Sat 11 Jan 18:34:29 GMT 2014, from the 1:7.8.0-0.68.20140110git1388931871 package. I tried again on a much older version of LMS running of an older version of Perl on a different machine and that worked as it should. Sorry, I don't know when this stopped working. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
Triode wrote: Can you confirm which skin and resolution you see this at? Is this the same for you Michael? 1280 x 720 pixels. The slow umlaut effect can be observed below in the two pictures. +---+ |Filename: Screen Shot 2014-01-10 at 9.02.29 PM.jpg | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=15469| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: H. Last time I looked at this, which admittedly was some time ago, start squeezelite with -v and it creates the /dev/shm/squeezelite file owner squeezelite, group squeezelite, mode rw-r--r--, so it was readable by anyone. Thought that jivelite user, (which the jivelite-autologin service is running as), just needs to read, rather than write the file. I'll take another look later today. I forgot about visulizer last time I updated the web-gui. Should have added a dedicated checkbox for that option. Oh well, at least you can put the -v in the options field for the moment. Edit: 20140109 12:50 Just had a quick look at code, squeezelite Code: vis_fd = shm_open(vis_shm_path, O_CREAT | O_RDWR, 0666); jivelite Code: vis_fd = shm_open(shm_path, O_RDWR, 0666); Still need to come back to this, just heading out of the front door right now. EDIT 20140109 13:30 Quick patch for squeezelite to get the shm file created 0666, by clearing the umask. I'll push a new build later. Code: --- squeezelite-1.5-c71ee71/output_vis.c.umask 2014-01-03 13:20:04.0 + +++ squeezelite-1.5-c71ee71/output_vis.c2014-01-09 13:26:28.224066710 + @@ -24,6 +24,7 @@ #if VISEXPORT +#include sys/stat.h #include sys/mman.h #include fcntl.h @@ -109,6 +110,8 @@ } void output_vis_init(log_level level, u8_t *mac) { + umask(000); + loglevel = level; sprintf(vis_shm_path, /squeezelite-%02x:%02x:%02x:%02x:%02x:%02x, mac[0], mac[1], mac[2], mac[3], mac[4], mac[5]); @@ -134,6 +137,8 @@ LOG_WARN(unable to open visualizer shared memory); vis_mmap = NULL; } + + umask(002); } #endif // VISEXPORT Thanks for looking at this. I tried your patch, but it doesn't seem to remove the requirement for squeezelite itself to have root to get a positive address for the shared memory pointer as tested in the next if statement. Code: if (vis_mmap 0) { pthread_rwlockattr_t attr; pthread_rwlocka In other words, I still see the following after patching and recompiling unless I use sudo with it: Code: [fedora@wandquad squeezelite]$ ./squeezelite -v -o plughw:CARD=UD501 -a 401 [07:38:46.523984] output_vis_init:137 unable to open visualizer shared memory Again, thanks for paying such quick attention, I know there are too many other things that demand your time right now! -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
JackOfAll wrote: dsdreamer, have you had any luck with getting the visulizer working, or are you still stuck with squeezelite complaining it cannot open the shm when running as non-root user? Did you try running it again as a system service, so it was running as the squeezelite user? I'm going to ask a stupid question. This is what we are talking about achieving, right? The dancing vertical bars, like a graphic equalizer display, is the visulizer, right? [image: http://www.communitysqueeze.org/files/jivelite_visulizer.png] I was kind of hoping if I left it late enough tonight, Triode would jump in, BANG, with the answer! ;) I've not seen any posts from him tonight. First of all, yes, having squeezelite being started by systemd is much more appropriate than starting it from an unprivileged command line. So that part of my problem was a phantom of my early morning stupor. Secondly, your patch does seem to help, now that I've tried it properly. Thirdly, yes I did mean exactly what you showed in you picture. Not to be outdone, I'll also show it now working: +---+ |Filename: Screen Shot 2014-01-09 at 8.00.04 PM.jpg | |Download: http://forums.slimdevices.com/attachment.php?attachmentid=15462| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
erland wrote: Sorry if this is a stupid question, but is there any kind of summary or overview release notes available for the Community Squeeze image that explain big changes in each version ? Related to this, is there some kind of very rough road map where it's possible to get a feeling what's planned to be done in the next couple of months or so ? As someone who doesn't follow this thread on daily basis it would be good to be able to get a feeling of the current state, some rough information about what has changed since the last image I installed and some indication about what's planned for next couple of months. Basically to get a better feeling when/if I should upgrade to a new version since I currently don't feel that I have the time to install and test each individual update immediately when they are published. I'm not asking for someone to write it in case it doesn't exist, I understand that in the current state the image might mostly be directed to people who have the time to follow this thread, I'm just asking if there already is some information available somewhere which I've missed. Erland, if you consider the user-space applications (Jivelite, Squeezelite and the Web-gui) to be the key to user-facing features, then a good overview of the past changes can easily be had here: https://code.google.com/p/squeezelite/source/list, here: https://code.google.com/p/jivelite/source/list, and here: https://github.com/CommunitySqueeze/Web-GUI/commits/master. But of course a lot goes on at the kernel and OS level to try to blend support for the Wandboard's devices with the mainline kernel, and to add support for particular audiophile USB DACs etc. I don't think we have a feature roadmap other than the broad intent as expressed here: http://www.communitysqueeze.org/faq.jsp . BTW, does IckStream have any forward looking manifestos or progress logs we can look at? -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Hello, I've been trying to get Jivelite to show visualizations while playing music on CSOS-F19-R2. I had been curious to see how the visualizations look, and had never been able to get them to work, except on my Fedora-20 laptop, where they worked just fine (except for the fact that jivelite wanted root to run when started from a non-priviledged account). I just now tried an experiment on my Wandboard Quad, where I started both squeezelite and jivelite as root, and was able to get the visualizations to work. I also noticed that running ./squeezelite -v produces an error message about not being able to open visualizer shared memory Code: [fedora@wandquad squeezelite]$ ./squeezelite -v -o hw:CARD=UD501 [22:31:13.619179] output_vis_init:134 unable to open visualizer shared memory Whereas if I use sudo in front of the same command, there is no such complaint, and a file is created under /dev/shm. I guess the real question is, how is this intended to work? Does the default installation of CSOS take care of permissions in such a way that the visualizer access to shared memory should work, or is there some other recommended way to make it so? -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
Triode wrote: I suspect that needs changes to the C code - nothing I have intentionally changed from squeezelplay. Look in jive_font.c and jive_label.c - at present it looks to be using the height of capital H to me - the font height is also the height of a capital + descending character. I suspect it needs something for ascending height too? Do you have a demo size and string where it is 100% reproducible? My example is Goldberg Variations, for keyboard (Clavier-Übung IV), where the umlaut on top of the capital U, gets left behind rather comically as the rest of the text scrolls away. I can send you a FLAC file of an offending track if you PM me a suitable email address. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
mherger wrote: Michael, I built SDL from scratch as well, but I configured it with ./configure --disable-video-x11, which along with commenting out line 94 of ./SDL-1.2.15/src/video/quartz/SDL_QuartzVideo.h, as hinted already in this thread Thanks, I'll give that a try! -- Michael In case you want to see the exact sequence I used, I've captured it in a simple shell script (attached as a zip file). It uses a patch file that I've included as well. This at least attempts to get the order correct, i.e., building the dependencies of the dependencies before the dependencies before jivelite... The patch is *only* for Mavericks, so don't use it for earlier OS X versions. +---+ |Filename: CGDirectPaletteRef_not_in_Mavericks.patch| |Download: http://forums.slimdevices.com/attachment.php?attachmentid=15451| +---+ -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
mherger wrote: Has anyone had success building jivelite on Mavericks? It did work on ML, but now it doesn't want to find X11/Xlib.h, or SDLmain. Tried to re-build SDL from scratch, but it fails Xlib.h as well, though it is there in /usr/local/include/X11 Michael, I built SDL from scratch as well, but I configured it with ./configure --disable-video-x11, which along with commenting out line 94 of ./SDL-1.2.15/src/video/quartz/SDL_QuartzVideo.h, as hinted already in this thread Code: //CGDirectPaletteRef palette;/* palette of an 8-bit display ...allows SDL to compile fine (except for some warnings about deprecated Cocoa backend API calls being used). I also had to compile a few dependencies and install them in /usr/local, using a typical routine of tar xvf something; cd something; ./configure; make; sudo make install, where something took turns at being each item in the following list. Code: LuaJIT-2.0.2 SDL-1.2.15 SDL_gfx-2.0.25 SDL_image-1.2.12 SDL_ttf-2.0.11 freetype-2.5.2 libpng-1.5.17 I had to use the older libpng than libpng-1.6.8, because the ./configure scripts for SDL_ttf-2.0.11 insisted on using that version for some of its test compiles and not the libpng-1.6.8, which I tried first. Finally, I made really small tweak to the one of the jivelite Makefiles: Code: diff --git a/src/Makefile.osx b/src/Makefile.osx index 42e2605..5d000ec 100644 --- a/src/Makefile.osx +++ b/src/Makefile.osx @@ -1,7 +1,7 @@ # makefile wrapper for osx # this is only needed if SDL is installed in /usr/local -CFLAGS = -I/usr/local/include/SDL +CFLAGS = -I/usr/local/include -I/usr/local/include/SDL LDFLAGS = -lSDLmain -lresolv -Wl,-framework,Cocoa Anyway, with those few build quibbles out if the way, I was able to build a functioning jivelite on Mac OS X and it seems to be working as expected. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Triode, On reflection, I think your dsd.c code does what it should, except for the belief that the actual number of sample_bytes per track for each channel must be even. When the source of the DSD is a DSF file rather than a DoP-encoded PCM file, there is no such constraint (the record companies had no concept of DoP encoding or its constraints when they authored the tracks). Thus the error trap as you have it is currently a bit trigger happy, and will give errors and warnings whenever there happens to be an odd number of sample_bytes, which should occur with 50% probability. If I am right in my analysis, you might want to silently accept the case of one sample byte left over, but still trap the zero case. Just to illustrate my meaning: Code: diff --git a/dsd.c b/dsd.c index 691d9cc..d1c9102 100644 --- a/dsd.c +++ b/dsd.c @@ -249,9 +249,18 @@ static decode_state _decode_dsf(void) { frames = min(bytes, d-sample_bytes) / bytes_per_frame; if (frames == 0) { - // /2 for dop should never result in 0 as header len is always even - LOG_WARN(frames got to zero); - return DECODE_ERROR; + if (d-sample_bytes == 1) { + LOG_INFO(had one Byte left over, insufficient to DOP encode); + LOG_INFO(sample bytes: FMT_u64, d-sample_bytes); + return DECODE_COMPLETE; + } + else { + // we really did get to zero, shouldn't happen + LOG_WARN(frames got to zero); + LOG_WARN(sample bytes: FMT_u64, d-sample_bytes); + LOG_WARN(bytes: %u, bytes); + return DECODE_ERROR; + } } frames = min(frames, out); -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Triode wrote: Yep - it was not handling the case of odd sized dsf files when playing with dop - please try latest git version which I hope fixes this. I have tested this and it works as expected, and I am enjoying the extra byte of reverb decay at the end of those odd-sized tracks! ;) -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
JackOfAll wrote: I just did a pull, build and quick test. So many things are barfed with regard to display on multiple skins, like not displaying the current playlist at all with Joggler skin, that I'm going to delete it from the testing repo and not announce it. ;) Which skin are you testing with? The HD1920? (Which is beyond the capability of my monitor.) Triode, I was able to pull the latest jivelite from git, build and take a look around. You have achieved what I was looking for in the HD Grid skin at 1920x1080p. The now playing list is back to a nice list format, and the repeat, and shuffle icons are now clearly visible below the track position slider. I certainly appreciate getting the visual cue that I may have selected a repeat or shuffle mode. I am less convinced that having these indications floating in a boardless area at the bottom of the screen is the most visually appealing choice that could be made. I have tried moving them around myself, and I find it difficult to make them look right whatever I do. Almost every UI I use (Windows, Mac OS X, most Linux Desktops) puts similar status icons in some kind of tray or panel. It is a visual language I am very used to, and have a hard time finding an alternative I really like. One possibe tweak I came up with was making the icons null for the repeat off and shuffle off states. This effectively hides the icons when they are not needed, which, given that I am have trouble coming to terms with any given positioning of them, is arguably a good thing. JackofAll is correct that this experiment has broken all of the other skins, even the non-grid HD skin in the same resolution doesn't display album lists correctly at the moment for me. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
JackOfAll wrote: I'll have to come back to this one later in the evening. Knee deep in generating kernel patch sets right now FYI: I did a Git pull yesterday and built from scratch on a Wandboard Quad. -Having uninstalled all the jivelite rpms that might conflict- (@JackofAll, I assume you did this too?), I then did a manual install from the just-built jivelite. Finally, I used: rm -rf ~jivelite/.jivelite/; systemctl restart autologin@tty1.service and went through the set-up selecting, the Joggler skin after selecting the language. No crash for me. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Announce: JiveLite - cut down squeezebox control application
Triode wrote: This probably needs more views from others and any input from a graphic artist/designer... I tried with the bar included but was not sure about this, especially as I'd had previous requests to remove it. I was wondering about a status box and trying to include file type or other information too. I personally hope you eventually do go in the direction of the status box. I know you have *many* other requests for your time, so I highly appreciate the effort made on my requests to date. You can always count on me to do testing and provide my feedback, but I know my limitations and am no visual GUI expert. I hope you have access to someone who can do this type of work for the long-term good of the project. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=98156 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix
Re: [SlimDevices: Unix] Community Squeeze OS F19 Release 1
Hi, I think I have the current latest of everything from the F19 arm testing repo. Code: community-squeeze-web.noarch1-61.20140102git7f01b95.fc19 @community-squeeze-testing dsdplay.armv7hnl0.1-6.20131229gitb4a052d.fc19 @community-squeeze-testing jivelite.armv7hnl 1:0.1-0.55.20140102gitd289907.fc19 @community-squeeze-testing logitechmediaserver.noarch 1:7.8.0-0.66.20131231git1387542508.fc19 @community-squeeze-testing squeezelite.armv7hnl1:1.5-0.2.20140101git80b66f6.fc19 @community-squeeze-testing That said, I have a recurrence of the previous problem whereby a pop-up message containing Can't open file for... appears at the end of most tracks, when playing back DSF albums. -- Dreamer, easy in the chair that really fits you... dsdreamer's Profile: http://forums.slimdevices.com/member.php?userid=12588 View this thread: http://forums.slimdevices.com/showthread.php?t=99395 ___ unix mailing list unix@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/unix