# Copyright 1999-2004 Gentoo Foundation # Distributed under the terms of the GNU General Public License v2 # $Header: /var/cvsroot/gentoo/xml/htdocs/proj/en/releng/release/2004.3/ChangeLog,v 1.9 2004/11/14 23:18:50 wolf31o2 Exp $ # This file serves as the official ChangeLog for Gentoo Linux 2004.3 # Each bugfix, enhancement, etc should be documented in this file # The syntax used by 'echangelog' should suffice. # PLEASE BE THOROUGH IN YOUR EXPLANATIONS! 14 Nov 2004; Chris Gianelloni : Closing bug #58512. Made sure ide-disk was built-in in the kernel configuration. 13 Nov 2004; Chris Gianelloni : Closing bug #61283. 2.5GHz G5 support was added to the ppc LiveCD. 13 Nov 2004; Chris Gianelloni : Closing bug #50978. The rp-pppoe package was added to livecd-stage1 for ppc. 13 Nov 2004; Chris Gianelloni : Closing bug #62909. Added ds21143 driver. 13 Nov 2004; Chris Gianelloni : Closing bug #61214. Added gcc and make to the 2004.3 ppc LiveCD (actually, just didn't remove them). 13 Nov 2004; Chris Gianelloni : Closing bug #56513. Added PReP support to LiveCD. 13 Nov 2004; Chris Gianelloni : Closing bug #49461. Added de21041 driver to LiveCD. 13 Nov 2004; Chris Gianelloni : Closing bug #30734. Added proper SCSI support for ppc. 13 Nov 2004; Chris Gianelloni : Closing bug #62707. A patched version of parted was added with proper HFS support. 13 Nov 2004; Chris Gianelloni : Closing bug #54894. A G3 kernel was added to the ppc LiveCD. 13 Nov 2004; Chris Gianelloni : Closing bug #53135. The lvm2 package was added to livecd-stage1. 13 Nov 2004; Chris Gianelloni : Closing bug #51949. Support was added for non-Apple SATA controllers. 13 Nov 2004; Chris Gianelloni : Closing bug #49549. Special care was taken to add better support for OldWorld Mac machines to the 2004.3 kernel configuration. 13 Nov 2004; Chris Gianelloni : Closing bug #43049. Added updated bootsplash and livecd-tools support for ppc. 22 Oct 2004; Chris Gianelloni : Closing bug #64486. Added slmodem to livecd-stage2. 22 Oct 2004; Chris Gianelloni : Closing bug #65097. Changed baselayout (as of 1.9.4-r6) to use /bin/bashlogin rather than mingetty's autologin hack. 22 Oct 2004; Chris Gianelloni : Removing depracated bootstrap-2.6.sh and bootstrap.sh. Renaming bootstrap-cascade.sh to bootstrap.sh. 22 Oct 2004; Chris Gianelloni : Closing bug #55378. Added scandelay option to add 10 second sleep to LiveCD. 22 Oct 2004; Chris Gianelloni : Closing bug #61628. Added dm-crypt support and it has been tested by tigger. 22 Oct 2004; Chris Gianelloni : Closing bug #54424. Well, we decided that this is a good idea, so we added it. The LiveCD now automatically asks for the keymap. After 10 seconds, or when the user hits enter, it defaults to *not* installing a keymap. This keeps us from breaking Mac and other clients. 22 Oct 2004; Chris Gianelloni : Closing bug #29667. This module would not build, so I added the sources to distfiles on the Universal CD in hopes that someone could use those to get online. I don't know if the modules just are not 2.6 compatable or what. 21 Oct 2004; Chris Gianelloni : Closing bug #67785. This was fixed by the xfsprogs maintainer when they made a newer version stable on supported arches. 21 Oct 2004; Chris Gianelloni : Closing bug #65792. This one was done upstream by the kernel developers once they released newer kernels with this driver. 21 Oct 2004; Chris Gianelloni : Closing bug #64760. This one was fixed by a patch to include all the 2.6 names for the usb modules in the usb module loading section of genkernel. 21 Oct 2004; Chris Gianelloni : Closing bug #64529. This was another case of me messing up on the last release. Anyway, since we built our current kernel configurations from the last release's 2.6 kernel configurations, this is already resolved. 21 Oct 2004; Chris Gianelloni : Closing bug #63091. While we did have evms on the LiveCD, we did not have the dm-bbr module. This has been added to the kernel specs and works, according to tigger. 21 Oct 2004; Chris Gianelloni : Closing bug #62529. This was simple enough as plasmaroo fixed it in genkernel, which means it was automatically done on any LiveCD that employs that version of genkernel or greater. 21 Oct 2004; Chris Gianelloni : Closing bug #62505. This one was fixed by dsd. All we had to do was include it in the kernel configuration file. 21 Oct 2004; Chris Gianelloni : Closing bug #60741. This was a very simple fix. All we had to do was put the contents of the handbook tarball under a handbook directory on the CD. 21 Oct 2004; Chris Gianelloni : Closing bug #62195. This was fixed by adding quotes around the variable names in the linuxrc script, which made them be processed as literal strings. This was a simple fix for a definite problem with the 2004.2 CD. 21 Oct 2004; Chris Gianelloni : Closing bug #61372. While I filed this bug myself, I ended up working it myself and let me say that I am definitely a lot closer to turning grey and bald. Anyway, genkernel/catalyst now support both bootsplash and gensplash. 21 Oct 2004; Chris Gianelloni : Closing bug #67764. While this one is not entirely a LiveCD bug, it is a package used on the LiveCD by default, so I'm including it here. Anyway, we've made kudzu-knoppix capable of build only libkudzu, which is all we really use on the LiveCD anyway. 21 Oct 2004; Chris Gianelloni : Closing bug #61610. This one was resolved by removing virtual/logger from the system profile, which was actually an accident on my part during the 2004.2 release cycle. 21 Oct 2004; Chris Gianelloni : Closing bug #61405. This one is simple enough. I set dhcpcd to run in the background rather than the foreground. This should improve boot times pretty dramatically on a machine that is not on a DHCP-enabled network. 21 Oct 2004; Chris Gianelloni : Closing bug #64514. This was fixed by plasmaroo in genkernel 3.0.2g, which has been in portage for a while. We are actually using genkernel 3.0.3, which also has this fix. 21 Oct 2004; Chris Gianelloni : Closing bug #60499. Here is another example of 2.6 prevailing where 2.4 fails. 21 Oct 2004; Chris Gianelloni : Closing bug #59938. We have the device-mapper stuff in the spec file and it has been tested by tigger, so if something goes wrong... blame him. 21 Oct 2004; Chris Gianelloni : Closing bug #59253. This is another one where the 2.6 kernel worked where the 2.4 kernel failed. Since we're all 2.6 now, this one is closed. 21 Oct 2004; Chris Gianelloni : Closing bug #58900. This was solved simply by switching to a 2.6-only setup for the 2004.3 LiveCD. 21 Oct 2004; Chris Gianelloni : Closing bug #58472. This was solved pretty simply by moving to a 2.6-only kernel for x86, which supports /proc/config.gz natively. 21 Oct 2004; Chris Gianelloni : Closing bug #58174. This one was reported for amd64, but affects all of us that have VIA chipsets. The problem was the lack of separation between libsata and IDE in the older kernels. This has been fixed upstream and is resolved. 21 Oct 2004; Chris Gianelloni : Closing bug #53378. We've verified that this is in the templates and barring any compilation problems, should be on every architecture that supports it. 21 Oct 2004; Chris Gianelloni : Closing bug #52717. This is another one that we didn't have to really do anything. Newer kernels have proper separation between the old and depracated IDE SATA interface and the newer and much better supported libsata version. 21 Oct 2004; Chris Gianelloni : Closing bug #51145. This was included in the 2.6.8 kernel, so we now have support for it. 21 Oct 2004; Chris Gianelloni : Closing bug #51086. Well, we made /usr/lib/hotplug/firmware tmpfs for catalyst. We also made /lib/firmware a symlink, in case there are any *broken* ebuilds that require this. Once the newer hotplug versions go stable, we'll swap which is which. There is also a tarball created of anything in /usr/lib/hotplug/firmware at the end of livecd-stage2. This way we do not lose any firmwares added by ebuilds. Oh yeah... blame seemant for getting this done. 20 Oct 2004; Chris Gianelloni : Closing bug #49516. There were some module problems with 2004.1 which was the cause for this bug being filed. I have ensured that ppp (both the module and the package) are available on the 2004.3 LiveCD, so this one is resolved. 20 Oct 2004; Chris Gianelloni : Closing bug #49328. This was the initial genkernel patch for udev support in the LiveCD/genkernel. Successful udev LiveCDs have been built on x86 and ppc64 using genkernel, so this one is definitely completed. 20 Oct 2004; Chris Gianelloni : Closing bug #47450. This was fixed by genkernel 3.0.2a, which issues an extra ramdisk_size option to the kernel for any initrd larger than 4096K in size. We also are not concerned with this on x86/amd64 since we have upped the default size for a ramdisk to 8192K. 20 Oct 2004; Chris Gianelloni : Closing bug #46975. We are including ide-floppy and floppy as modules, so this one is definitely fixed. 20 Oct 2004; Chris Gianelloni : Closing bug #46297. This one was fixed in baselayout 1.9.4-r3. Since we are using 1.9.4-r5, this one is no longer going to affect us. 20 Oct 2004; Chris Gianelloni : Closing bug #44091. This was accomplished by extending genkernel to check if the lvm2 package is installed and if it was compiled statically. If it was, then the proper binaries are copied into the initrd. You can blame tigger and plasmaroo for this one. 20 Oct 2004; Chris Gianelloni : Closing bug #44045. The new LiveCD uses a 2.6.9 kernel, which has greater support for SATA devices by making a distinct separation between IDE and SATA. Since we are only using the libsata interface, this bug is no longer valid on the newer CD. 20 Oct 2004; Chris Gianelloni : Closing bug #16202. Added globespan-adsl to packagecd.spec in grp/src/packages for the Universal LiveCD. I have also added it to livecd-stage2.spec in boot/kernel/gentoo/packages, so if it builds against 2.6.9, there will be a ready-to-use module on the LiveCD, also.