This file documents the instructions for upgrading to Slackware 13.0, the packages added, removed, renamed, and/or split during the development cycle from Slackware 12.2 through 13.0, and some potential "gotchas" that users can avoid by arming themselves with a little knowledge. *** INSTRUCTIONS FOR UPGRADING FROM 12.2 *** Follow the instructions detailed in the UPGRADE.TXT located in this directory. Note that upgrading from a Slackware version earlier than 12.2 is NOT supported at all and will most likely not work. Also note that upgrading from 12.2 to 13.0 (64bit) is not supported. *** PACKAGE ADDITIONS SINCE 12.2 *** a/xz ap/cupsddk ap/dc3dd d/swig kde/guidance-power-manager kde/kaudiocreator kde/kdebase-runtime kde/kdebase-workspace kde/kdepimlibs kde/kdeplasma-addons kde/kdevelop kde/kdevplatform kde/konq-plugins kde/ktorrent (moved from /extra) kde/mplayerthumbs kde/skanlite l/M2Crypto l/PyQt l/QScintilla l/akonadi l/automoc4 l/babl l/boost l/chmlib l/clucene l/djvulibre l/eigen2 l/exiv2 l/gegl l/gmm l/ilmbase l/iso-codes l/jasper l/libarchive l/libdvdread l/libical l/libmcrypt l/libmsn l/libnl l/libspectre l/libvncserver l/libxklavier l/libzip l/openexr l/phonon l/pil l/pycups l/qca-cyrus-sasl l/qca-gnupg l/qca-ossl l/qimageblitz l/qtscriptgenerator l/raptor l/rasqal l/redland l/seamonkey-solibs l/sip l/soprano l/strigi l/taglib-extras l/urwid n/crda n/iw n/iwlwifi-5150-ucode n/snownews x/intel-gpu-tools x/scim-input-pad x/xcb-util x/xorg-server-xephyr x/xpyb x/xscope xap/MPlayer xap/electricsheep xap/thunar-volman xap/xfce4-power-manager extra/kde3-compat/arts extra/kde3-compat/kdelibs3 extra/kde3-compat/qca-tls1 extra/kde3-compat/qca1 extra/kde3-compat/qt3 extra/xf86-video-intel-alternate/xf86-video-intel-* (several alternate versions of the Xorg intel driver just in case the default doesn't work properly for you) *** PACKAGE REMOVALS SINCE 12.2 *** kde/kdeaddons kde/knemo l/arts l/beecrypt l/dbus-qt3 l/qca-tls l/svgalib_helper (won't build on 2.6.28+ kernels) n/bitchx x/liboldX x/xf86-input-calcomp x/xf86-input-digitaledge x/xf86-input-dmc x/xf86-input-dynapro x/xf86-input-elo2300 x/xf86-input-jamstudio x/xf86-input-magellan x/xf86-input-magictouch x/xf86-input-microtouch x/xf86-input-palmax x/xf86-input-spaceorb x/xf86-input-summa x/xf86-input-tek4957 x/xf86-input-ur98 x/xf86-video-cyrix x/xf86-video-fbdev x/xf86-video-i810 (use xf86-video-intel instead) x/xf86-video-imstt x/xf86-video-nsc x/xf86-video-vga x/xf86-video-via (use xf86-video-openchrome instead) x/xphelloworld x/xrx xap/gimp-help-2 *** OTHER NOTABLE CHANGES AND HINTS *** l/dbus moved to a/dbus New system user accounts: oprofile (uid=51) New system group accounts: oprofile (gid=51) dialout (gid=16) netdev (gid=86) The Slackware installer now uses udev to initialize your hardware, including the network interface card(s). This has positive consequences for network installations (using NFS, FTP, HTTP or SMB). You no longer have to run the 'pcmcia' and 'network' scripts prior to running 'setup' - the network interface will be created and intialized by udev. If a DHCP server is found on your local network, the setup program will let you choose between automatic configuration of your network interface using DHCP or specifying a static IP address. Using udev, the commandline for fully unattended configuration and startup of the dropbear SSH server has changed slightly. Suppose you want to boot the 'hugesmp' kernel, use DHCP for interface eth0, and you have a us-english keyboard layout: the commandline to auto-start the SSH daemon in the installer would become: hugesmp.s kbd=us nic=auto:eth0:dhcp Note: if you do not want to use udev, the "auto" keyword in that example commandline must be replaced with the actual name of the network module for your card. If you do not want to use udev, you must add the parameter "noudev" to the command line that boots the Slackware installer, and the original ("old") Slackware hardware configuration scripts will be used. Use one of the provided generic kernels for daily use. Do not report bugs until/unless you have reproduced them using one of the stock generic kernels. You will need to create an initrd in order to boot the generic kernels - see /boot/README.initrd for instructions. The huge kernels are primarily intended as "installer" and "emergency" kernels in case you forget to make an initrd. For most systems, you should use the generic SMP kernel if it will run, even if your system is not SMP-capable. Some newer hardware needs the local APIC enabled in the SMP kernel, and theoretically there should not be a performance penalty with using the SMP-capable kernel on a uniprocessor machine, as the SMP kernel tests for this and makes necessary adjustments. Furthermore, the kernel sources shipped with Slackware are configured for SMP usage, so you won't have to modify those to build external modules (such as NVidia or ATI proprietary drivers) if you use the SMP kernel. If you decide to use one of the non-SMP kernels, you will need to follow the instructions in /extra/linux-2.6.29.6-nosmp-sdk/README.TXT to modify your kernel sources for non-SMP usage. Note that this only applies if you are using the Slackware-provided non-SMP kernel - if you build a custom kernel, the symlinks at /lib/modules/$(uname -r)/{build,source} will point to the correct kernel source so long as you don't (re)move it. As usual, there are changes in udev packaging that need mentioning... As with 12.2, the system udev rules now reside in /lib/udev/rules.d/ instead of /etc/udev/rules.d/ in older versions. There should never be a reason to edit anything in /lib/udev/rules.d/, so if you think you have a case where this is required, either you're wrong or it needs to be addressed in the upstream source. However, you can override default rules by placing one with an identical name inside /etc/udev/rules.d/ The rules files in /etc/udev/rules.d/ are still intended to (maybe) be edited as needed by local system administrators, and as such, the rules for optical and network devices will still be placed there. Also, be sure to have the new dialout group added to your system, or udev will kindly remind you in the system logs... Due to the upgrade of kde from 3.5.10 to 4.2.4, you will need to move your existing $HOME/.kde/ out of the way (either completely remove it or back it up somewhere else if you think you might need it again for whatever reason); otherwise, you will almost surely experience odd configuration problems with kde applications. If mailto: links don't work properly (or at all) in firefox, you may have to remove an existing $HOME/.mozilla/mimeTypes.rdf file and restart Firefox. If you are unable to access the cups configuration web interface from your browser, you may have to blacklist the ipv6 module and reboot. This is not an acceptable solution, of course, but it's the only one we have at the moment. HP multifunction printer/scanners require that your user account be a member of the "lp" group for hp-toolbox to work properly, and to use the scanner portion of some (all?) units, you'll need to be a member of the "lp" group. This is because hplip's udev rules set the device with group "lp" ownership. HAL is not new anymore, but here are a few notes related to it: 1. User accounts with permission to mount removable devices must be in at least the "plugdev" group. 2. User accounts with permission to do power-management tasks, such as suspend, hibernate, reboot, and shutdown, via HAL methods should be in the "power" group. 3. HAL will honor settings in /etc/fstab if a device is present there, so you could technically have removable devices defined in /etc/fstab, but if the fstab settings do not allow normal users to mount them (with the "user" or "users" option), then HAL/dbus will not allow them to be mounted either. In other words, for example, if your fstab line for the cdrom/dvd drive includes the "owner" option, you will not be able to mount it as a normal user. 4. If you find a need for modified fdi files, those should be placed in the relevant directories in /etc/hal/ instead of /usr/share/hal/ If you notice Xfce's Terminal and perhaps some other applications being drawn very slowly in X, then you should try explicitly disabling the Composite extension in /etc/X11/xorg.conf, or set XLIB_SKIP_ARGB_VISUALS=1 in your environment prior to starting X. For more information on this, see: http://bugzilla.xfce.org/show_bug.cgi?id=2792 We've also gotten a report of some other things (such as VirtualBox) that might benefit from this. Speaking of Xorg, the version of Xorg shipped with Slackware 13.0 will not (in most cases) require an /etc/X11/xorg.conf file at all. Configuration of input devices and such is handled by HAL, and the X server autoconfigures everything else. You can still create an xorg.conf file if you wish, or you can create a minimal xorg.conf with only the specific contents that you wish to override (as an example, to use a binary-only video driver). Due to removed drivers and other such changes, it's quite possible that your old xorg.conf will not work correctly with this version of Xorg. If you need to use a non-US keyboard layout, then copy the file located at /usr/share/hal/fdi/policy/10osvendor/10-keymap.fdi to /etc/hal/fdi/policy and edit it to suit your needs. Have a look at the contents of that file for an example and more information. If you prefer to do this the "old" way using /etc/X11/xorg.conf, then you can use "X -configure" or "xorgsetup" to generate an xorg.conf, then add the following lines to the "ServerFlags" section to disable input device hotplugging via HAL: Option "AllowEmptyInput" "false" Option "AutoAddDevices" "false" Option "AutoEnableDevices" "false" This is also relevant if you prefer to disable HAL completely for whatever reason. If you are using input hotplugging via HAL and a synaptics touchpad, then you might need to copy /usr/share/hal/fdi/policy/11-x11-synaptics.fdi to /etc/hal/fdi/policy/ and edit it to suit your needs. You can also use synclient(1) to make changes "on the fly." If you want to try the new kernel mode setting (KMS), you don't have to build a custom kernel; add this to your kernel's lilo stanza: append = "i915.modeset=1" If you are using a KVM switch, you might experience problems with the mouse when switching from one system to another. If so, you probably need to be using the imps protocol for the psmouse driver, and that's a simple edit: uncomment the following line in /etc/modprobe.d/psmouse.conf: #options psmouse proto=imps Next, unload and reload the psmouse module (do this as root): modprobe -r psmouse ; modprobe psmouse If you have set up an encrypted root partition, you will need to have access to your keyboard in order to type the passphrase. This may require you to add the uhci-hcd and usbhid modules to your initrd image if you have a USB keyboard. Also note that if you are using a non-US keyboard, you can use the '-l' parameter to the 'mkinitrd' command in order to add support for this keyboard to your initrd. If you have permission errors when attempting to burn a cdrom or dvd image, such as the following: /usr/bin/cdrecord: Operation not permitted. Cannot send SCSI cmd via ioctl then cdrecord almost certainly needs root privileges to work correctly. One potential solution is to make the cdrecord and cdrdao binaries suid root, but this has possible security implications. The safest way to do that is to make those binaries suid root, owned by a specific group, and executable by only root and members of that group. For most people, the example below will be sufficient (but adjust as desired depending on your specific needs): chown root:cdrom /usr/bin/cdrecord /usr/bin/cdrdao chmod 4750 /usr/bin/cdrecord /usr/bin/cdrdao If you don't want all members of the 'cdrom' group to be able to execute the two suid binaries, then create a special group (such as 'burning' which is recommended by k3b), use it instead of 'cdrom' in the line above, and add to it only the users you wish to have access to cdrecord and cdrdao. If you have compilation errors that look something like this: /usr/include/asm-generic/fcntl.h:117: error: redefinition of 'struct flock' /usr/include/bits/fcntl.h:142: error: previous definition of 'struct flock' /usr/include/asm-generic/fcntl.h:140: error: redefinition of 'struct flock64' /usr/include/bits/fcntl.h:157: error: previous definition of 'struct flock64' See the following link for some pointers on fixing it: http://www.mail-archive.com/blfs-dev@linuxfromscratch.org/msg08942.html Input methods for complex characters (CJK, which is shorthand for Chinese, Japanese, Korean) and other non-latin character sets have been added. These input methods use the SCIM (Smart Common Input Method) platform. The environment variables for SCIM support are set in /etc/profile.d/scim.sh The requirements for getting SCIM input methods to work in your X session are as follows: (1) Use a UTF-8 locale. Look in /etc/profile.d/lang.sh for setting your language to (for instance) en_US.UTF-8. As a word of warning: maybe you should leave root with a non-UTF-8 locale because you don't want root's commands to be misinterpreted. You can add the following line to your ~/.profile file to enable UTF-8 just for yourself: export LANG=en_US.UTF-8 (2) Make the scim profile scripts executable. These will setup your environment correctly for the use of scim with X applications. Run: chmod +x /etc/profile.d/scim.* (3) Start the scim daemon as soon as your X session starts. The scim daemon must be active before any of your X applications. In KDE, you can add a shell script to the ~/.kde/Autostart folder that runs the command "scim -d". In XFCE you can add "scim -d" to the Autostarted Applications. If you boot your computer in runlevel 4 (the graphical XDM/KDM login) you can simply add the line "scim -d" to your ~/.xprofile file. This gives you a Desktop Environment independent way of starting scim. When scim is running, you will see a small keyboard icon in your system tray. Right-click it to enter SCIM Setup. In 'Global Setup' select your keyboard layout, and you are ready to start entering just about any language characters you wish! Press the magical key combo in order to activate or deactivate SCIM input. The SCIM taskbar in the desktop's corner allows you to select a language. As you type, SCIM will show an overview of applicable character glyphs (if you are inputting complex characters like Japanese). If you are using the pinentry-gtk2 interface (for entering passphrases with gpg-agent), be aware that there is a bug in the way scim-bridge and the pinentry-gtk2 interact. The result is that keyboard input does not register with pinentry-gtk2. For the time being, either change the /usr/bin/pinentry symlink to use the qt or curses frontend, or don't use scim. If you have an older machine (with a BIOS released prior to 2001) and it will not power off on shutdown, try adding this to your kernel's lilo stanza: append = "acpi=force" If you have a Dell Optiplex 760 (or perhaps some other machine that has the newer ICH10 chipset), and it won't boot, try one of these: 1. Update the BIOS image to at least A03 2. Turn C-States control off in the BIOS 3. Boot with "hpet=disabled" If your wireless and/or bluetooth radios are not turned on by default after booting up, you might need to load the rfkill-input module. If that solves the problem for you, edit /etc/rc.d/rc.modules.local or /etc/rc.d/rc.local to load the module on boot, or create /etc/modprobe.d/rfkill.conf and put the following line in it: install rfkill /sbin/modprobe -i rfkill ; /sbin/modprobe -i rfkill-input