| Commit message (Collapse) | Author | Age |
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38340
|
|
|
|
|
|
|
|
| |
This is needed to make modprobe work
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38020
|
|
|
|
|
|
| |
symbols do not exist in the config
SVN-Revision: 37400
|
|
|
|
|
|
|
|
| |
the procd changes broke the postinst insmod.
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 37015
|
|
|
|
|
|
|
|
| |
Makes warnings/errors visible when building with V=w/V=1.
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 35713
|
|
|
|
|
|
| |
location
SVN-Revision: 34804
|
|
|
|
| |
SVN-Revision: 34715
|
|
|
|
| |
SVN-Revision: 34683
|
|
|
|
|
|
| |
dependencies
SVN-Revision: 34450
|
|
|
|
|
|
| |
build_dir/linux-* directories when building different libc implementations in the same buildroot
SVN-Revision: 34299
|
|
|
|
| |
SVN-Revision: 34228
|
|
|
|
| |
SVN-Revision: 34225
|
|
|
|
|
|
| |
incorperate it into the kernel metapackage version, make kmods depend on this specific version. The aim of this change is to invalidate kmods which are built against a different kernel config on the opkg metadata level. Manual copying and insmod of custom *.ko files, e.g. for development purpose, is still possible.
SVN-Revision: 29686
|
|
|
|
|
|
|
| |
modules.builtin is only available in 2.6.33 and later, so fall back to
assuming it's built-in for older kernels.
SVN-Revision: 29018
|
|
|
|
|
|
| |
kmods
SVN-Revision: 28856
|
|
|
|
| |
SVN-Revision: 28650
|
|
|
|
| |
SVN-Revision: 28649
|
|
|
|
| |
SVN-Revision: 28627
|
|
|
|
| |
SVN-Revision: 28360
|
|
|
|
| |
SVN-Revision: 27939
|
|
|
|
|
|
| |
missing-fhonour-copts warnings with external toolchains
SVN-Revision: 27771
|
|
|
|
|
|
|
| |
Linux 3.0 calls itself 3.0.0, so insmod expects the modules in
/lib/modules/3.0.0.
SVN-Revision: 27761
|
|
|
|
|
|
|
|
|
| |
- 4 field kernel versions
implement CompareKernelPatchVer using this new helper
implement kernel_patchver_{gt,ge,eq,le,lt}, which are more convenient for use with inline $(if)
SVN-Revision: 27087
|
|
|
|
| |
SVN-Revision: 27032
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When doing a build with tracing on, the expansion of ModuleAutoLoad can get a
little hairy.
Using intermediate variables to name the arguments makes tracing more readable.
One side effect is that if an argument is accidentally left out, we won't get
all of the parameters shifted one over thanks to quoting (done in AutoLoad).
Signed-of-by: Philip Prindeville <philipp@redfish-solutions.com>
SVN-Revision: 23513
|
|
|
|
|
|
| |
only where necessary
SVN-Revision: 22720
|
|
|
|
|
|
| |
modules in ./packages
SVN-Revision: 22146
|
|
|
|
| |
SVN-Revision: 22145
|
|
|
|
| |
SVN-Revision: 20569
|
|
|
|
|
|
| |
required for mounting the rootfs (patch by cshore)
SVN-Revision: 19878
|
|
|
|
| |
SVN-Revision: 18777
|
|
|
|
| |
SVN-Revision: 18301
|
|
|
|
| |
SVN-Revision: 18295
|
|
|
|
|
|
| |
board name around in kmod-* package versions anymore
SVN-Revision: 17528
|
|
|
|
| |
SVN-Revision: 17293
|
|
|
|
| |
SVN-Revision: 15249
|
|
|
|
| |
SVN-Revision: 14741
|
|
|
|
| |
SVN-Revision: 14739
|
|
|
|
| |
SVN-Revision: 14738
|
|
|
|
| |
SVN-Revision: 13986
|
|
|
|
|
|
| |
now applies to *ALL* kernel versions, and is copied along with ./files-* - this gets rid of quite a bit of redundancy in the extra kernel drivers.
SVN-Revision: 13010
|
|
|
|
|
|
| |
kernel dependent packages
SVN-Revision: 12503
|
|
|
|
| |
SVN-Revision: 12414
|
|
|
|
|
|
| |
config passthrough target which will evaluate the original config in the generated package. This is done so kernel modules can have configuration menus like busybox, etc... Signed-off-by: Axel Gembe <ago@bastart.eu.org>
SVN-Revision: 11139
|
|
|
|
|
|
| |
watchdog changes in 2.6.24+ kernels (closes: #3402)
SVN-Revision: 11115
|
|
|
|
| |
SVN-Revision: 10972
|
|
|
|
| |
SVN-Revision: 9640
|
|
|
|
|
|
| |
compile issues)
SVN-Revision: 9635
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If KERNEL_CC isn't set, we end up with a "CC=" on the kernel build
command-line. We don't always need CC, as the CROSS_COMPILE flag does
the job instead. In fact, specifying CC messes up the build when we're
using a biarch compiler.
This change doesn't specify CC= if the KERNEL_CC variable is empty.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
SVN-Revision: 9408
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
'powerpc' is a valid arch for the LINUX_KARCH variable, but the build
system unconditionally replaces LIUNUX_KARCH=powerpc with
LINUX_KARCH=ppc.
This change only does the replacement if LINUX_KARCH isn't set. This
allows us to use the powerpc architecture.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
SVN-Revision: 9405
|