Jo-Philipp Wich [Thu, 19 Jan 2012 15:37:47 +0000 (15:37 +0000)]
base-files: do not pass --set-worker /lib/hotplug2/worker_fork.so to hotplug2 anymore, its statically linked - patch by Viktar Palstsiuk <viktar.palstsiuk@promwad.com>
Jo-Philipp Wich [Wed, 18 Jan 2012 23:46:06 +0000 (23:46 +0000)]
ext-toolchain.sh: add option to write .config Add a new --config option to generate a .config from a given toolchain, this allows for easy integration of external toolchains, e.g.
Jo-Philipp Wich [Wed, 18 Jan 2012 19:22:14 +0000 (19:22 +0000)]
mtd: external toolchains usually do not know about MTDREFRESH, so search in $(LINUX_DIR)/include first, fixes compilation with external CodeSourcery toolchain
Jo-Philipp Wich [Wed, 18 Jan 2012 03:32:30 +0000 (03:32 +0000)]
ext-toolchain.sh: - use GCC's own idea of the target name (-dumpmachine) - display help if called without args - add copyright foo - remove some leftover flag handling code
Jo-Philipp Wich [Wed, 18 Jan 2012 03:15:22 +0000 (03:15 +0000)]
invoke patch-specs.sh in the gcc/final install stage. Patch up OpenWrts own cross toolchains to always search headers and libraries in $STAGING_DIR, this should solve most issues with missing headers, indirect linking and not found libraries. At a later stage, all -I and -L flags will be purged from TARGET_LDFLAGS and TARGET_CPPFLAGS.
Jo-Philipp Wich [Wed, 18 Jan 2012 03:11:45 +0000 (03:11 +0000)]
add patch-specs.sh, a utility for modifying GCC specs The patch-specs.sh utility dumps the GCC specs of a given toolchain and modifies them to always include $STAGING_DIR in the link and compiler command lines, this makes most -I and -L flags unnecessary and lets the compiler automatically find libraries and headers in the staging dir, also solves the majority of -rpath issues.
Jo-Philipp Wich [Wed, 18 Jan 2012 03:08:09 +0000 (03:08 +0000)]
use ext-toolchain.sh to integrate external toolchains Use ext-toolchain.sh to wrap external toolchain commands, abort build if certain features such as CONFIG_SOFT_FLOAT or CONFIG_IPV6 are enabled but not supported by the toolchain.
Jo-Philipp Wich [Wed, 18 Jan 2012 03:03:24 +0000 (03:03 +0000)]
add ext-toolchain.sh, a utility for external toolchains The ext-toolchain.sh utility script implements various external toolchain tasks: * Testing of available features, like c++, soft-float or ipv6 support * Finding the libc implementation (uclibc or eglibc/glibc) * Inferring the GNU target name * Finding shared objects for packaging (libc, libpthread, ...) * Finding executables for packaging (ldd, gdbserver, ...) * Generating wrapper scripts for toolchain commands
John Crispin [Tue, 17 Jan 2012 12:10:35 +0000 (12:10 +0000)]
gigaset 76x
* fixes leds
* adds all leds found on the sx763 board
* removes board_config from flash layout because wireless data is stored in kernel (ath5k)
* fix button handling
Signed-off-by: Luka Perkov <openwrt@lukaperkov.net>
SVN-Revision: 29763
John Crispin [Tue, 17 Jan 2012 11:23:11 +0000 (11:23 +0000)]
ramips: Rework ramips_eth to not require irqsave locking anymore
Previously the tx housekeeping was done in a spin_lock_irqsave critical
section which causes irqs to be disabled during that time. Since the
housekeeping is already prepared to be scheduled as a tasklet process
the housekeeping only in softirq context and revise the locking between
the tx path and the housekeeping tasklet by using a normal spin_lock
which in most situations will be a NOP anyway.
This makes sure that interrupts are only disabled for a short time
since in the worst case the housekeeping might have to free up to 256
skbs.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
SVN-Revision: 29762
Hauke Mehrtens [Fri, 13 Jan 2012 20:11:23 +0000 (20:11 +0000)]
brcm47xx: update usb driver
Add workarround for BCM5357 and BCM4749.
This patch contains the following fixes from George Kashperko:
* separate subroutines for controller initialization workarounds -
following CodingStyle recommendations;
* __devinit __devexit __devinitconst annotations for CONFIG_HOTPLUG;
* fix leak in ssb_hcd_create_pdev/bcma_hcd_create_pdev - as hci_res being
kmalloc'ed is never freed anywhere while platform_device_add_resources
will kmemdup resources right away;
* fix compilation error in ssb_hcd_resume - it will fail as soon as
CONFIG_PM is selected.
Felix Fietkau [Thu, 12 Jan 2012 10:15:11 +0000 (10:15 +0000)]
ar8216: use vlan configuration for connecting ports (forced to be untagged) even with vlan_enable=0, but keep them isolated before the switch has been configured this makes enable_vlan=0 practical for vlan passthrough
Daniel Dickinson [Wed, 11 Jan 2012 13:47:48 +0000 (13:47 +0000)]
Remove invalid LED GPIO from GW6200 board definition. GPIO 35 isn't intended to control ethernet LED (the led behaviour is a side effect) and using the gpio as an led causes kernel panics.
Jo-Philipp Wich [Sun, 8 Jan 2012 15:41:34 +0000 (15:41 +0000)]
pptp: remove support for underlying ipproto, it is adviced since a while that pptp should be declared as separate interface alognside wan - in this configuration the ipproto will cause stray udhcpc errors due to not existent interfaces, so remove it
Jo-Philipp Wich [Sun, 8 Jan 2012 15:34:23 +0000 (15:34 +0000)]
introduce a USE_EXTERNAL_LIBC symbol which gets selected by EXTERNAL_TOOLCHAIN || NATIVE_TOOLCHAIN Currently we always assume uClibc if an external toolchain is used, this breaks for non-uClibc toolchains or even vanilla uClibc ones since they do not share the external librpc semantics as OpenWrt. Solve the problem by defining an abstract "EXTERNAL_LIBC" which packages might or might not depend on.
Jo-Philipp Wich [Sun, 8 Jan 2012 15:23:11 +0000 (15:23 +0000)]
include: calculate md5sum over sorted kernel config symbol list and 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.
Gabor Juhos [Sat, 7 Jan 2012 19:36:33 +0000 (19:36 +0000)]
ar71xx: allow to specify max read size for m25p80
Reading from the flash chip on the TL-WR2543ND seems buggy.
If the SPI flash driver tries to read too much data in one
SPI transfer, the flash chip returns bogus values. This can
be caused by a buggy flash chip on my board, or it can
be a bug in our SPI driver.
Add a workaround to the m25p80 driver until I find out the
root cause of the problem. The patch allows to specify the
maximum numner of bytes which can be read safely withint
one SPI transfer.
Gabor Juhos [Sat, 7 Jan 2012 17:24:44 +0000 (17:24 +0000)]
ar71xx: local mac support on the wndr3700
The wndr3700 at least has no eth0 mac address and usually leverages
the first wireless device's mac when in a bridged scenario. If,
however, you want to route, and not bridge the interfaces, you
need a unique mac address for it.
This patch sets the local bit on the mac address pulled from the
wireless chip and uses the resulting address for eth0.
Patch-by: Dave Taht <dave.taht@gmail.com>
SVN-Revision: 29675
Gabor Juhos [Sat, 7 Jan 2012 16:29:16 +0000 (16:29 +0000)]
generic: improve MIPS kexec support
It is based on patches from the linux-longsoon-community git tree:
http://dev.lemote.com/cgit/linux-loongson-community.git/
Now the kernel can use the command line parameter from kexec-tools.
Runtime tested on ar71xx with 2.6.39.4 (the wathdog must be stopped
before executing the new kernel). Compile tested with lantiq (3.1.4)
and brcm47xx (3.0.12).
projects / thirdparty / openwrt.git / log
