| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
| |
This patch adds support for TP-Link TL-WA7210n[0]
[0] http://www.tplink.com/en/products/details/?model=TL-WA7210N
Signed-off-by: Hendrik Frenzel <hfrenzel@scunc.net>
SVN-Revision: 44526
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44525
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44523
|
|
|
|
|
|
| |
Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44522
|
|
|
|
|
|
|
|
|
|
|
| |
This adds the GMAC entries in the ipq806x dtsi file as well as in the
ap148 specific dts file.
This also adds the MDIO change as well.
Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44520
|
|
|
|
|
|
| |
Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44519
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This change is fixing the issues observed when booting from NOR flash
with SMP enabled.
Error logs below:
building xattr subsystem, 0 of xdatum (0 unchecked, 0 orphan) and 0 of
xref (0 dead, 0 orphan) found.
m25p80 spi32766.0: SPI transfer failed: -110
spi_master spi32766: failed to transfer one message from queue
jffs2: Write of 873 bytes at 0x019001e8 failed. returned 0, retlen 792
Patch is cherry-picked from here:
https://www.codeaurora.org/cgit/quic/qsdk/oss/kernel/linux-msm/commit/drivers/spi/spi-qup.c?h=coconut_20140924&id=4faba89e3ffbb1c5f6232651375b9b3212b50f02
More details in the patch file.
Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44518
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It took very little to make the TL-WDR4900 work under 3.19:
- config is the same as for 3.18
- only patch 210 had to be refreshed, the other patches are
the same as for 3.18
- in /etc/config/wireless the path options need to be prefixed
with "platform/" ('platform/ffe09000.pci/...')
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
SVN-Revision: 44517
|
|
|
|
|
|
| |
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 44514
|
|
|
|
|
|
| |
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44513
|
|
|
|
|
|
| |
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 44511
|
|
|
|
|
|
|
|
|
|
| |
This sets 'enable_vlan' and thus uses 802.1q
VLANs, but without tagging on either interface.
Signed-off-by: Claudio Leite <leitec@staticky.com>
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 44509
|
|
|
|
|
|
|
|
|
| |
Toggles the LA bit on the WAN interface.
Signed-off-by: Claudio Leite <leitec@staticky.com>
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 44508
|
|
|
|
|
|
|
|
|
| |
ipq806x target has been upgraded to 3.14. There is no need to maintain
this file anymore.
Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44505
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The watchdog driver already registers a restart notifier, we just have
to enable it in the config and in the DT to fix the "reboot" command.
This is done by integratin the following patch-set:
https://lkml.org/lkml/2015/2/20/610
I'm copy-pasting the description below:
qcom-wdt is currently assuming the presence of a dedicated node in DT
to gets its configuration. However, on msm architecture, the watchdog is
usually part of the timer block. So this patch-set is changing the driver
and slightly enhancing the timer DT bindings to provide the relevant clocks
and interrupts.
Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44504
|
|
|
|
|
|
|
|
|
|
| |
Some bootloaders seem to trigger the watchdog during the boot process,
therefore the lack of watchdog driver trigger a reboot a few seconds
after boot. So we'll enable it here to avoid it.
Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44503
|
|
|
|
|
|
|
|
| |
Updated patch that fixes whitespace and supports kernel v3.3 too.
Signed-off-by: Owen Kirby <osk@exegin.com>
SVN-Revision: 44502
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
MPIC timers are enabled in the config, however a devicetree clock setting
is missing causing the following error:
/soc@ffe00000/timer@41100: cannot get timer frequency.
/soc@ffe00000/timer@42100: cannot get timer frequency.
This patch adds the missing clock and avoids the error.
It's a functional copy of this code
do_fixup_by_compat_u32(blob, "fsl,mpic",
"clock-frequency", get_bus_freq(0), 1);
in arch/powerpc/cpu/mpc85xx/fdt.c in the u-boot code.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
SVN-Revision: 44501
|
|
|
|
|
|
|
|
|
| |
Defaulting to disabled, based on the value of
the (old?) CONFIG_RALINK_USBPHY
Signed-off-by: Claudio Leite <leitec@staticky.com>
SVN-Revision: 44500
|
|
|
|
|
|
|
|
|
|
|
| |
For Nexx WT3020 board_name is wt3020, not nexx-wt3020
root@OpenWrt:~# . /lib/ramips.sh; echo $(ramips_board_name)
wt3020
Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
SVN-Revision: 44499
|
|
|
|
|
|
|
|
|
| |
Size 0x160000 was added to support Buffalo WZR-600DHP2, however Netgear
devices (R6250, R8000) require 0x180000.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 44497
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44496
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44495
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44494
|
|
|
|
|
|
| |
Signed-off-by: Owen Kirby <osk@exegin.com>
SVN-Revision: 44492
|
|
|
|
|
|
| |
Signed-off-by: Owen Kirby <osk@exegin.com>
SVN-Revision: 44491
|
|
|
|
|
|
| |
Signed-off-by: Owen Kirby <osk@exegin.com>
SVN-Revision: 44490
|
|
|
|
|
|
| |
Signed-off-by: Your Name <blogic@openwrt.org>
SVN-Revision: 44489
|
|
|
|
|
|
|
|
| |
Thanks to Bruno Randolf for testing and Manuel Lauss for providing a fix.
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44487
|
|
|
|
|
|
|
|
|
|
|
|
| |
The firware md5sum check never worked due to missing dollar signs.
Also suppress
`md5sum: can't open '/lib/firmware/ath10k/QCA988X/hw2.0/firmware-3.bin'`
error on any board not using ath10k.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 44486
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44485
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44482
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44481
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I got work with this patch.
Ethernet switch (includes VLAN), WiFi connected via PCIe, LEDs, buttons.
In mtd partion map of DTS file, I renamed Linux firmware regions (kernel + root squashfs) to “firmware”because it allows kernel to split kernel and roots and rootfsdata.
signed-off-by: ngc@ff.iij4u.or.jp
——
——
SVN-Revision: 44470
|
|
|
|
|
|
|
|
|
|
|
| |
To choose whether bulid or not RT288x PCIe bus driver, CONFIG_SOC_RT2880 is wrong.
Here is its fix, and enables PCIe bus driver for some targets which have 2nd WiFi chipset via PCIe bus.
signed-off-by: ngc@ff.iij4u.or.jp
————
————
SVN-Revision: 44469
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- according to imx6 Makefile and u-Boot documentation is itb
and probably should not be changed
- this fixes build error if CONFIG_TARGET_ROOTFS_INCLUDE_FIT is set
(missing .itb file)
- use DTS_DIR (like in imx6 Makefile)
only compile tested
Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
Acked-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 44468
|
|
|
|
|
|
| |
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 44467
|
|
|
|
|
|
| |
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 44464
|
|
|
|
|
|
| |
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 44463
|
|
|
|
|
|
| |
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 44462
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44459
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44458
|
|
|
|
|
|
| |
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44456
|
|
|
|
|
|
|
|
|
|
| |
The current pxa3xx_nand driver doesn't support sub-page writing,
so the VID header offset must be specified explicitly.
Signed-off-by: Claudio Leite <leitec@staticky.com>
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 44455
|
|
|
|
|
|
|
|
| |
oops, accidently moved instead of copied them
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44454
|
|
|
|
|
|
|
|
| |
ZyXEL NBG6716 "rfkill" is a switch, not a button.
Signed-off-by: Marcin Mikolajczak <gr4ffy at gmail.com>
SVN-Revision: 44453
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some Lantiq SoCs are not able to use buffered writes properly with
Intel command set flash due to the way NOR addresses on EBU are
manipulated. This patch disables buffered writes on those devices.
The only device affected at the moment is ARV4510PW, others use
AMD/Fujitsu command set.
Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
SVN-Revision: 44451
|
|
|
|
|
|
|
|
|
| |
For targets with NO_XIP ltq_mtd->map[i].phys equals -1 and devm_ioremap fails.
Fix this by using pdev->resource[i].start instead.
Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
SVN-Revision: 44450
|
|
|
|
|
|
| |
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 44449
|
|
|
|
|
|
|
|
|
|
| |
console
procd gets last console from cmdline, so swapping tty1 (framebuffer) and ttyAMA0 (uart) is enough to force procd output via uart
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 44448
|