Does ALSA depend on HAL?

Hello,
recently I decided to get rid of HAL and uninstalled it from my system. Later the same day I suddenly discovered that sound stopped working. I thought that it might be because of HAL, so I reinstalled it, stopped the alsa daemon, started the hal daemon, the started the alsa daemon again, but the sound did not come back. Later the same day I did a full system upgrade (which didn't seem to contain any kernel-related stuff, mostly xorg updates) and rebooted. I was surprised to see that the sound started working again.
So, my question is, what the hell was that?! :-) Does ALSA depend on HAL in any way? Or did the full upgrade bring my sound back? If yes, why did it stop working?

Maybe you could compare the outputs of lsmod for both situations. Maybe there are some modules missing or something like that.
Other question, how do you "measure" that alsa doesn't work? Do you simply try to play something with some program? I'd recommend to start alsamixer in a terminal and see if it works. If so, the modules are up and running.
By the way, did you know that the alsa daemon simply launches alsactl store and restore? It only saves your currently set volumes when it stops and reloads them when it starts. So a restart of the alsa daemon is simply useless. That's a common mistake many arch users make.

Similar Messages

  • External USB HD does not work with HAL:

    External USB HD does not work with HAL:
    From the last 10 days the HAL auto-mount with ntfs-3g file system of my external hard drive  does not woks.
    But the manual mount procedure with the command mount -t ntfs-3g /dev/sdb* /mnt/XYZ works.
    I've followed the instructions in the wiki: http://wiki.archlinux.org/index.php/HAL and the system has worked up to the last kernel or HAL update.
    My fat32 usb-stick works without any problem.
    If I try to open the device with KDE it appears an error popup wit this message:
    TODO: have to rethink extra options
    Error: kio_media_mount_helper
    The log from the kernel.log is
    Nov 6 22:03:16 myhost usb 5-1: new high speed USB device using ehci_hcd and address 4
    Nov 6 22:03:16 myhost usb 5-1: configuration #1 chosen from 1 choice
    Nov 6 22:03:17 myhost Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
    Nov 6 22:03:17 myhost ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
    Nov 6 22:03:17 myhost Initializing USB Mass Storage driver...
    Nov 6 22:03:17 myhost scsi6 : SCSI emulation for USB Mass Storage devices
    Nov 6 22:03:17 myhost usb-storage: device found at 4
    Nov 6 22:03:17 myhost usb-storage: waiting for device to settle before scanning
    Nov 6 22:03:17 myhost usbcore: registered new interface driver usb-storage
    Nov 6 22:03:17 myhost USB Mass Storage support registered.
    Nov 6 22:03:22 myhost scsi 6:0:0:0: Direct-Access WDC WD32 00JB-00KFA0 0811 PQ: 0 ANSI: 0
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] 625142448 512-byte hardware sectors (320073 MB)
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] Test WP failed, assume Write Enabled
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] Assuming drive cache: write through
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] 625142448 512-byte hardware sectors (320073 MB)
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] Test WP failed, assume Write Enabled
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] Assuming drive cache: write through
    Nov 6 22:03:22 myhost sdb: sdb1
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] Attached SCSI disk
    Nov 6 22:03:22 myhost sd 6:0:0:0: Attached scsi generic sg2 type 0
    Nov 6 22:03:22 myhost usb-storage: device scan complete
    and the error.log is
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] Assuming drive cache: write through
    Nov 6 22:03:22 myhost sd 6:0:0:0: [sdb] Assuming drive cache: write through
    Do you have some idea ....
    Bye.

    I have done the following:
    Downloaded the patched hal version.
    used makepkg in /abs/local/trunk/ to create a pkg for pacman
    used sudo pacman -U pkgname to install it.
    used sudo nano /etc/hal/fdi/policy/preferences.fdi to create the fdi file with the following content:
    saved the file and removed my ntfs entryes from fstab and rebooted. and still there is no automount off ntfs partitions.
    have i done something wrong?
    br fjodork
    WARNING: what follows is a guide of what did work for me, I'm still in an early stage of learning linux, so please correct me if anything of what follows is wrong/dangerous/stupid. As I said, this worked for me, it's not pretty, since I was learning each step on the fly and I'm sure there are better ways to do this.
    So fjodork, you seem to have done the the necessary steps, though i couldn't get it to work until i first removed hal without dependencies
    pacman -Rd hal
    it will give you some warning about the dependencies, ignore  it since we are reinstalling hal in a few steps.
    First i tried to follow Raymano's link, obviously it showed me nothing in the browser so I thought it might be a broken link.
    So I downloaded the source tarball from here http://hal.freedesktop.org/releases/hal-0.5.10.tar.gz
    unpacked it changed the /tools/hal-storage-mount.c as per Milfadoodle's instructions.
    at first I tried to compile directly from source, the process would finish without a glitch though after reboot nothing changed, as if hal weren't
    installed, i guess there were issues with paths. anyway read on
    Then repacked the source in tarball, copied it to /var/abs/extra/system/hal and modified the PKGBUILD as follows
    makedepends=('pkgconfig' 'gperf')
    options=('!libtool')
    install=hal.install
    source=(${pkgname}-${pkgver}.tar.gz #removed url so makepkg looks in the current directory for the tarball
        hal
        hal-policy.patch
        cryptsetup_location.patch
        hal-0.5.9-hide-diagnostic.patch
        ntfs3g-valid-options.patch)
    md5sums=('6641c30a27c00485c6accac5110ff911' # replace with md5sum of your package's sourceball (run md5sum filename)
             '882f67668cb14a0a9e4a27ef22278027'
             '5ba8b610aa9763a5f42b9f7cbd7a86ad'
    then makepkg (i had to add --asroot, don't ask why
    and finally pacman -U hal-0.5.10-1-i686.pkg
    another reboot and everything worked without a glitch!:D (yeah i know i still haven't got rid of the reboot mentality )
    I have also put the policy files as per the wiki instructions.
    PS: I know this is a dirty way to getting things done but as i said, I'm just starting
    If anyone could comment/correct/explain this post I'd be more than happy, the automount issue was bothering me for about a month now (you'll say why do I need ntfs in the first place? well, I only recently switched for Linux as my primary OS, these are some relics of the M$ era

  • Why does KOffice depend on gtk?

    When I try to install KOffice, I get:
    resolving dependencies... done.
    looking for inter-conflicts... done.
    Targets: libgsf-1.14.3-2  wv2-0.2.3-1  lcms-1.16-1  libwmf-0.2.8.4-4
             atk-1.18.0-2  cairo-1.4.6-2  libthai-0.1.8-1  pango-1.16.4-1
             gtk2-2.10.13-1  libcroco-0.6.1-1  librsvg-2.16.1-1
             imagemagick-6.3.4.4-1  python-2.5.1-1  libwpd-0.8.9-1
             libexif-0.6.16-1  koffice-1.6.3-2  libdatrie-0.1.1-1
    Total Package Size:   49.97 MB
    Proceed with installation? [Y/n]
    Why does it depend on gtk2?
    Last edited by miggols99 (2007-07-08 17:30:57)

    librsvg is dependent on GTK2. This was discussed at KDE-devel:
    http://lists.kde.org/?l=koffice-devel&m … 425620&w=2
    Last edited by skottish (2007-07-08 17:44:57)

  • If I buy a ipad mini with cellular, and not sign up for cellular service, can I use the onboard GPS, or does it depend on the cell signal since it is aGPS

    If I buy a ipad mini with cellular, and not sign up for cellular service, can I use the onboard GPS, or does it depend on the cell signal since it is aGPS?

    Right. But as I understand it, the aGPS is a function that speeds location using cell signals, but isn't necessary for the GPS signal. I can buy a plugin device that is a real GPS that I can use with my iPad mini. However, I would need an app for maps. I was wondering if anyone had used one of these.

  • Why does jdk depend on jre ?

    Installing jdk through pacman calls jre as a dependancy, however, the jdk packages shouldn't require jre (the functionality is included with jdk).
    Note:
    JDK 6 Update 2
    The Java SE Development Kit (JDK) includes the Java Runtime Environment (JRE) and command-line development tools that are useful for developing applets and applications.
    http://java.sun.com/javase/downloads/index.jsp
    and
    Contents of the JDK
    Runtime Environment
    (In the jre/ subdirectory) An implementation of the Java Runtime Environment (JRETM) for use by the JDK. The JRE includes a JavaTM Virtual Machine (JVMTM), class libraries, and other files that support the execution of programs written in the JavaTM programming language.
    http://java.sun.com/javase/6/webnotes/README.html
    In the past, on other distro's I've never installed JRE when I wanted JDK. Is this an error on the maintainers part or...?
    Last edited by steven (2007-07-25 10:45:04)

    Cerebral wrote:
    steven wrote:
    It's simple, the JDK has the runtime environment built in, so there's no need to install both the jdk and the jre. If someone is in the scenario you describe whereby, they have JRE already installed and decide to install the JDK afterwards, then pacman should be able to detect the existing jre and simply remove the runtime environment from the JDK package (which currently happens). I'm simply saying, if someone only wants JDK, don't install 2 seperate packages as the JDK includes JRE anyway.
    Surely in the PKGBUILD or install files in pacman, it can make checks for a pre-existing JRE install (when installing JDK) and only remove the runtime from the JDK, but if there's no pre-existing JRE install, then ONLY install JDK, without removing the runtime.
    My Internet connection is super, so I don't have a big issue with bandwidth, but surely it's better for your mirrors and users who don't have super connections. It also just makes more sense. It seems completely daft to download 60 megs of 2 packages, then delete jre from the jdk and install it seperately. It's bizarre.
    I don't think you understand how our packaging system works... the jdk we provide, that users download from our repos, does not have the "included" jre - users don't download the jre twice.  In your suggestion, they would download the jre twice (once when they install jre, and once again later when they install jdk).
    Besides that matter, it is not a simple thing to dynamically alter the contents of packages at install time - we provide binary packages, pre-built, and the file list of a package is essentially exactly equal to the contents of the tarball you download.
    The way we have it now, only one package in our repos contains the jre, and that's the jre package.  I don't see why you'd rather your package manager sees one package instead of two - you have no duplicate files on your system, and downloaded no extra data (since the binary jdk package doesn't contain the jre at all).
    Ah ok I see, I had the wrong idea.
    AndyRTR wrote:check our jdk PKGBUILD and you will find a line   rm -rf linux-jdk/jre
    This quote gave me the idea that it's download, but then the built in functionality is removed before it's actually installed. Are you saying that this isn't the case?
    shining wrote:You're so good that you don't even need to run the programs you write
    I wish

  • Why does vim depend on vi?

    Running a very minimal machine in minimal space, I have a custom vim package that strips out gui, perl, python and ruby intergration because i don't use any of those features. i was thinking of also removing vi to save try to save the 28 megabytes (it is important, its a 2 gig usb stick), since vi and vim come from the same source and I initially though they really shouldn't need to depend on each other.
    However, it turns out that vi creates the /usr/share/vim folder which both it and vim use and need. This looks like the only reason vim needs vi, and obviously its done this way to prevent package file conflicts. The 28 IS pretty much the size of the /usr/share/vim folder, so its a stupid space "optimization" to persue, as without vi, vim would just be that much bigger itself, but for curiosity's sake, is it possible to build vim without vi? I couldn't figure out what trick causes the vi PKGBUILD to include it and the vim one to drop it. What causes vi package to create a /usr/share/vim and prevents the vim package from doing the same?

    Also, vi lacks X support and vim is compilled --with-features=big while vi only has --with-features=normal.
    Vi, for example, won't do 256 colors.

  • Does Firefox depend on xulrunner? Should it?

    Hi,
    I just got a broken firefox after upgrading to ff28.
    I've seen this before, ff complains that it cannot find XPCOM.
    I solved it by upgrading manually xulrunner. I'd have thought this would mean that ff depends on xulrunner, doesn't it?
    Cheers,
    Charlie

    No.
    expac -Q '%D' firefox
    gtk2 mozilla-common libxt startup-notification mime-types dbus-glib alsa-lib desktop-file-utils hicolor-icon-theme libvpx libevent nss hunspell sqlite
    Did you do a full upgrade when firefox version was bumped?

  • Does xulrunner depend on GNOME or not?

    Hi... I have two questions (the rest is just me talking about how confused I am):
    1. How is mozilla getting its files to use shared libraries even when they aren't linked against those shared libraries?
    2. How can I tell if it's safe to delete a shared library that is used in only this way?
    =====
    I know xulrunner comes with two files in /usr/lib/xulrunner-1.9.2/components that bring in gnome dependencies: libmozgnome.so and libnkgnomevfs.so. Firefox also comes with libnkgnomevfs.so but according to this bug https://bugzilla.mozilla.org/show_bug.cgi?id=512671 it's not supposed to.
    I'm wondering how these libraries actually get used.
    [root@connors-laptop firefox-3.6]# cd /usr/lib/firefox-3.6
    [root@connors-laptop firefox-3.6]# ldd firefox | grep gnome
    [root@connors-laptop firefox-3.6]# cd /usr/lib/xulrunner-1.9.2/
    [root@connors-laptop xulrunner-1.9.2]# ldd libxul.so | grep gnome
    [root@connors-laptop xulrunner-1.9.2]# ldd libxpcom.so | grep gnome
    [root@connors-laptop xulrunner-1.9.2]# ldd libmozjs.so | grep gnome
    [root@connors-laptop xulrunner-1.9.2]#
    As far as I can tell... NOTHING is linked against them. However the situation gets even weirder. I know that libxul and libxpcom are essential for firefox, yet firefox seems not to be linked against them either:
    [root@connors-laptop firefox-3.6]# ldd firefox | grep libxul
    [root@connors-laptop firefox-3.6]# ldd firefox | grep libxpcom
    [root@connors-laptop firefox-3.6]# grep libxul firefox
    Binary file firefox matches
    [root@connors-laptop firefox-3.6]# grep libxpcom firefox
    Binary file firefox matches
    [root@connors-laptop firefox-3.6]#
    The binary seems to contain references to them but they are runtime references, not compile time references? It dynamically decides whether or not it should use those libraries when you start it? I tried figuring out where the gnome libraries are used this way and I think I found it:
    [root@connors-laptop firefox-3.6]# cd /usr/lib/xulrunner-1.9.2/
    [root@connors-laptop xulrunner-1.9.2]# grep gnome libxul.so
    Binary file libxul.so matches
    Thunderbird is causing similar confusion with me. It is of course independent from xulrunner but it comes with a libmozgnome.so and I can't tell how it's used. Anecdotally everyone says it's safe to delete libmozgnome and libnkgnomevfs but how do I know this? If I deleted orphan libraries that things aren't linked against I would delete libxul, libxpcom and libmozjs which is obviously NOT okay to do.

    Ok so when I start the code... it is "deciding" whether to be linked against libmozgnome.so or not. Sounds a bit like self-modifying code to me. I was wondering because if I made a program that used gconf for example, even if I wanted it to be "optional" and put it in a section of the program that I considered "unimportant"... if I didn't have gconf installed I would absolutely get an undefined reference to symbol and nothing would work... even if I think it's "not a big deal."
    I guess this LD_PRELOAD stuff is the solution. Firefox can first check if the right gnome deps are installed. If they are, it will load them at run time, if they are missing it will just move on. I guess this is how openoffice uses the "optional" gconf. It comes with a file called gconfbe1.uno.so but nothing is linked (at compile time) to it. This must mean it is loaded while openoffice runs because openoffice detected at runtime that I have gconf but if I didn't it would've skipped this dependency.
    So ldd is not a good way to check if a program does this. I should use lsof to see if a program is dynamically loading libraries? And if I want to confirm my suspicions about how it does so I should grep the executable for phrases like "LD_PRELOAD", "LD_LIBRARY_PATH" or "dlopen"? I guess the ultimate way would be checking the source code if I have time.

  • Why do I have cc and cc2014 both in my list of apps in creative cloud? Can I delete the older and keep the newer or does one depend on the other?

    I have a list of both CC and CC2014 in all my programs on my Creative Cloud 'dashboard'.
    Do I need to keep them all, do they depend on each other, or can I delete the CC and just keep 2014?
    PS: I have never yet found the CSS panel where you can choose different settings in Dreamweaver CC and CC2014 - it looks entirely different from CS6. Can you tell me where to locate that or is it gone?

    http://blogs.adobe.com/jkost/2014/06/installing-the-2014-release-of-creative-cloud.html
    This messages says (at least some) CC 2014 programs use NEW plugins https://forums.adobe.com/thread/1499663
    -so do not uninstall the older CC programs if you use plugins in your programs until you are sure you have plugins that work in CC2014
    If you are sure you don't need the old CC programs
    -http://helpx.adobe.com/creative-cloud/help/install-apps.html to install or uninstall
    -read reply #3 about the ORDER of uninstalling & installing https://forums.adobe.com/thread/1242671
    For program specific questions you need to ask in that forum
    If you will start at the Forums Index https://forums.adobe.com/welcome
    You will be able to select a forum for the specific Adobe product(s) you use
    Click the "down arrow" symbol on the right (where it says All communities) to open the drop down list and scroll

  • Does the dependent task/milestones across projects in a Master project get updated automatically?

    Hi
    We have a master project which has 5 sub projects in our PS 2013 environment. The milestones in the sub projects have dependency across projects with in the Master project.
    Would like to know, if we update a milestone in one of the sub projects, would it automatically update the other milestone in another project (which has dependency)..?
    Please suggest.

    When you open the sub-project, you can see the corresponding linked tasks in grey color. As soon as you modify, can see the changes in the respective linked tasks.
    Later when you open the other sub project, you will get a prompt like "Links between Projects 'project name'"
    choose the task and accept the changes, later click on close.
    Then you can see the changes reflected in the respective tasks.
    Not mandatory to open the master project and expand.
    Cheers. Happy troubleshooting !!! Sriram E - MSFT Enterprise Project Management

  • How do I set an achor point in Illustrator cs 6 on PC? Does it depend on the tool I'm using?

    I'm trying to fill in some missing color in a small spot on a hat and I may need to set an achor point. I want to use the blob brush though it is to big. What tool what a use to segregate a specific setting of the hat?

    RC,
    That depends on the shape of the area to have the colour. You could use a circle, maybe.
    If there is nothing there, you could also create a path larger than the area, give it the desired colour, and dragging it behind the (rest of the) hat, which you can do in the Layers palette.
    But depending on the artwork, you may have better options.
    And depending on the version, you may be able to just use Live Paint to fill the gap.

  • Does CRC depends on MTU size ?

    Hello everybody,
    Can someone please confirm the following :
    1) Whether Switching in Physical layer casues CRC on router interface for DS3 TDM link.
    2) Also whether Increment in CRC depends on MTU size on router interface.
    Thanks

    Hello!
    1) Yes this will cause CRC errors.
    2) I do not think this will make a difference
    Mike

  • [SOLVED] Does E17 Depend on Qt or Gtk ?

    E17 will be released about the end of this year. It looks very cool. I want to try it, but I don't know whether it needs any other heavy libraries like GTK or QT ? Anyone knows ?
    Last edited by ShinChyn (2009-11-05 01:57:12)

    E17 has it's own menus, but it's using gnome-menus for it's XDG back end. The actual menu specification is from freedesktop.org. But since gnome-menus was designed for Gnome, its developers naturally made the gnome-icon-theme its default. It's not actually a dependency though. One can easily use many other icon themes and can even remove gnome-icon-theme totally.
    I believe that all of this is happening because the devs at Enlightenment assume that if one is using Linux as a desktop/workstation, they probably already have GTK2 installed. It's a fairy safe assumption. I rarely hear of anyone that's using a purely QT based system, but there are many like myself that only use GTK2 and terminal based apps.

  • Iphoto, does it depend on the hardware or software version?

    I bought a second hand mac, but it didnt came with iLife and I didnt get the discs. So I need buy iPhoto.
    I´m told the new iPhoto doesnt work with old macs. Not sure if they refferr to old software or hardware.
    I have OS X 10.6.8 (cant even get Lion on this mac)  in a macbook white with just 1g ram.
    can anybody help on wich version of iPhoto can I use? I´ll later have to figure out where to get it.
    thanks

    If your Mac meets these requirements you'll be able to use iPhoto 11 on your Mac:
    An Intel dual-core processor
    Mac OS X 10.6.3 or newer (known as “Snow Leopard”)
    At least 1GB of RAM
    At least 5GB of free hard-drive space
    OT

  • [Solved] Fresh install; HAL won't start?

    So I just finished a fresh installation and now I can't start X because HAL won't start up at all.
    I haven't messed around with anything except for installing everything and fixing up my config files.
    Seems like HAL is crashing because of permissions but I even checked the paths and they seem to be fine:
    hald --daemon=no --verbose=yes
    01:08:41.090 [i] hald.c:673: hal 0.5.14
    01:08:41.090 [i] hald.c:674: using child timeout 250s
    01:08:41.090 [i] hald.c:739: Will not daemonize
    01:08:41.090 [i] hald_dbus.c:5444: local server is listening at unix:abstract=/var/run/hald/dbus-uBDySnOMjt,guid=33e68e9ab5cc00ac1ce4ba92000008c1
    01:08:41.091 [i] hald_runner.c:304: Runner has pid 1695
    Runner started - allowed paths are '/usr/lib/hal:/usr/lib/hal/scripts:/usr/bin'
    01:08:41.093 [i] hald_runner.c:184: runner connection is 0x17da830
    01:08:41.097 [W] osspec.c:393: Unable to open /proc/mdstat: No such file or directory
    ** (process:1694): WARNING **: Failed to add monitor on '/usr/share/hal/fdi/preprobe': Permission denied
    ** (process:1694): WARNING **: Failed to add monitor on '/etc/hal/fdi/preprobe': Permission denied
    ** (process:1694): WARNING **: Failed to add monitor on '/usr/share/hal/fdi/information': Permission denied
    ** (process:1694): WARNING **: Failed to add monitor on '/etc/hal/fdi/information': Permission denied
    ** (process:1694): WARNING **: Failed to add monitor on '/usr/share/hal/fdi/policy': Permission denied
    ** (process:1694): WARNING **: Failed to add monitor on '/etc/hal/fdi/policy': Permission denied
    22:08:41.098 [i] mmap_cache.c:126: Regenerating fdi cache..
    Could not get password database information for UID of current process: User "???" unknown or no memory to allocate password entry
    Could not get password database information for UID of current process: User "???" unknown or no memory to allocate password entry
    22:08:41.099 [E] hald_runner.c:888: Error running 'hald-generate-fdi-cache': org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.
    22:08:41.099 [i] mmap_cache.c:104: In regen_cache_cb exit_type=2, return_code=0
    22:08:41.099 [E] mmap_cache.c:153: fdi cache regeneration failed!
    22:08:41.099 [i] mmap_cache.c:156: fdi cache generation done
    22:08:41.099 [i] mmap_cache.c:278: cache mtime is 0
    *** [DIE] osspec.c:osspec_init():449 : Unable to read /proc/mounts
    Can't find anything worthwhile in logs etc., and I can't think of anything that could be causing this?
    I checked that the HAL user exists and has been added to the default groups, but that "???"-part is alarming.
    DBus starts up normally.
    Networking and "lo" seem to be working fine as well, and I don't have iptables up.
    Last edited by initbox (2011-07-05 16:59:36)

    tomk wrote:Update your system - xorg does not depend on hal anymore. hal is obsolete.
    Ah, haven't been up to date on all the changes, it seems...
    In the past I've had problems with the nvidia driver + Xorg without HAL, I think?
    Anyhow, contrary to the Xorg page, I had no 20-nvidia.conf. I created that and now I'm able to startx without HAL but KDM doesn't work, the screen flickers when it attempts to start Xorg but then it crashes. Strange. Nothing in the Xorg logs either.
    I'll have to see about that next, but it's nice to be able to get rid of HAL at least.

Maybe you are looking for