Philipp Zabel [Sat, 4 Feb 2006 21:02:13 +0000 (21:02 +0000)]
update-modules: fix module loading during first boot
- Phil's patch from http://www.handhelds.org/hypermail/oe/current/6167.html
- this also works when ipkg intercepts update-modules in postinst.
Justin Patrin [Sat, 4 Feb 2006 01:49:45 +0000 (01:49 +0000)]
efl, e17: Update to newer release tarballs
- Removed no-splash patch for e-wm, crash bug seems to be gone
- add "released" entrance
- remove ecore fixes (fixed in release...finally)
- add new configure fix for evas and change configure patch names to use ${PV}
Rene Wagner [Fri, 3 Feb 2006 21:54:04 +0000 (21:54 +0000)]
matchbox-common: remove "Utilities" category as the distinction between "Utilities" and "Other" is currently rather random. Part of a fix for hh.org Bug #1462
Justin Patrin [Fri, 3 Feb 2006 17:27:02 +0000 (17:27 +0000)]
meta-e-x11, task-e-x11: move task-e-x11* to task-e-x11.bb and create meta packages to conform with new standard
- Also removed meta-e as it's not needed
John Bowler [Fri, 3 Feb 2006 02:28:21 +0000 (02:28 +0000)]
busybox: fix udhcpc with multiple ports on slugos in 1.01
- a new 'simple.script' patch ensures that resolve.conf is
maintained correctly when multiple ethernet ports are present
in the system, fixes slug bug 347
Matthias Goebl [Thu, 2 Feb 2006 17:29:32 +0000 (17:29 +0000)]
packages/lirc: Update lirc to 0.8.0
- Update 0.7.2+0.8.0pre4 to 0.8.0
- Inherit module-base - class for compilation of kernel-module
- Start lircd and lircexec only if config exists
- Describe sub-packages
- Use oe-provided pre/post-scripts
busybox 1.01: smaller config for OpenZaurus
- removed Minix FS support
- removed DevFS support
- removed DHCP server
- removed unzip (we have it in separate package)
- removed busybox keymaps handling support
- switched buffer allocation policy to "Allocate on the Stack" like Familiar have
Philipp Zabel [Wed, 1 Feb 2006 20:46:00 +0000 (20:46 +0000)]
udev: add 084
- Fixes the severe udev initial boot speed issues for hx4700.
- DEFAULT_PREFERENCE="-1", please try if it works for you.
- Uses udevsynthesize to generate initial hotplug events.
John Bowler [Wed, 1 Feb 2006 08:25:30 +0000 (08:25 +0000)]
slugos-init: add init script to handle USB device moving in 0.10
- the init.d script fixfstab uses blkid to build a partition/uuid
mapping table and compares this to the version from the previous
boot, if anything changes /etc/fstab is updated
John Bowler [Wed, 1 Feb 2006 02:05:17 +0000 (02:05 +0000)]
binutils: make strip and objcopy return 0 if asked to write ro files in 2.16
- the same bug exists in earlier versions and is probably fixed by the
same patch, objcopy (==strip) would identify an ro file and abort the
overwrite (smart_rename), but the caller in objcopy fails to check the
error code and so the error is not reported
Matthias Hentges [Tue, 31 Jan 2006 23:52:30 +0000 (23:52 +0000)]
altboot:
- More kernel 2.6 changes
- Booting off SD seems to work as expected now
- Turn off printk before showing the menu, kernel messages make a mess of it
- Load keympas before showing the menu. init=/bin/sh is useless with a b0rked keymapping.
Koen Kooi [Tue, 31 Jan 2006 21:58:34 +0000 (21:58 +0000)]
h3900: stop people thwarting my attempt to get thumb-interwork running
people without a h3900 machine should also stop messing with this file, damnit!s
Rene Wagner [Tue, 31 Jan 2006 20:44:21 +0000 (20:44 +0000)]
ppp: bump PR to force rebuild after package.bbclass changes. only add postinst for ppp (main) package. Fixes hh.org Bug #1469.
- no misc changes this time.
Rene Wagner [Tue, 31 Jan 2006 20:24:32 +0000 (20:24 +0000)]
h3900.conf: don't include tune-xscale.conf so as to retain backwards compatibility. the extra bit of performance isn't worth the hassle of building, storing/mirrorin, and maintaining duplicate packages.
John Bowler [Tue, 31 Jan 2006 17:36:37 +0000 (17:36 +0000)]
package.bbclass: implement a failsafe strip in classes
- package.bbclass now uses file-native and cross strip to reliably strip
unstripped executables and check the return code. For the moment a
failure here doesn't cause the build to fail but does output a failure
message direct to the controlling terminal (this is temporary). This
behaviour can be changed by forcing IGNORE_STRIP_ERRORS = ""
The change works by using a new runstrip shell function for every
potentially strippable file. Since this calls native 'file' there is
a dependency on all inheritors of package.bbclass for file-native and
since this would break native builds (because of the cycle in
file-native and it's own native depends) native.bbclass cancels this
dependency (system file should be fine for native packages and, anyway
native currently doesn't package.)
John Bowler [Tue, 31 Jan 2006 17:13:37 +0000 (17:13 +0000)]
file-native: fix do_stage in 4.13
- the hand crafted do_stage installed a broken file in staging, the
fix is to use the standard native do_stage which does a proper install
using the makefile
John Bowler [Tue, 31 Jan 2006 09:29:38 +0000 (09:29 +0000)]
slugos-init: make /linuxrc use disk uuids, remove .recovery stuff in 0.10
- the /boot/disk bootstrap now mounts -U uuid if one is available from
turnup. turnup checks disk for blkids and no longer complains about
.recovery, /boo/* no longer create .recovery or check for it on boot;
disconnect the disk or the network to boot to flash on a semi broken
system.
John Bowler [Tue, 31 Jan 2006 09:25:12 +0000 (09:25 +0000)]
e2fsprogs: split out blkid for use in initrd in 1.38
- blkid (and uuidgen) are useful for non-ext2/3 file systems
(e.g. reiserfs) and can be used at boot to reliably find the
root file system, therefore it is useful to have them in
separate packages. e2fsprogs (${PN}) now RDEPENDS on the new
packages to ensure nothing gets broken.
Matthias Hentges [Mon, 30 Jan 2006 10:53:54 +0000 (10:53 +0000)]
altboot:
- First changes to support Kernel 2.6
- As of this push altboot should be considered *unstable* until work on 2.6 has finished!
- Add IGNORE_STRIP_ERRORS = "1" to allow for snapshots
John Bowler [Mon, 30 Jan 2006 09:27:28 +0000 (09:27 +0000)]
devlabel: release 0.48.01
- this does actually seem to work (release 0.45.01 simply exited from
bash without any output), but it depends on things like the 'raw'
command existing and working, so it isn't very useful
Justin Patrin [Mon, 30 Jan 2006 08:18:38 +0000 (08:18 +0000)]
efl, e17: update CVS versions again, fix cpp issues
- evas is left at 20060113 because it changed that day and breaks ecore
- cpp and gcc paths are hard-coded to /usr/bin/ to fix some issues with OE's binaries in use with edje
John Bowler [Mon, 30 Jan 2006 03:05:50 +0000 (03:05 +0000)]
linux: fix missing HOST_CC_ARCH in module compilation in kernel/module all
- define TARGET_KERNEL_CC_ARCH to define arch specific compiler options
in machine.conf. The default for the kernel is "" (empty), although
it should probably be TARGET_CC_ARCH in the future. The default for
modules is TARGET_CC_ARCH