The initialization routines for these boards were relying on some (wrong)
defaults for the QCA953x ethernet. Make these defaults explicit to prevent
breaking them when the QCA953x defaults are fixed.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r46206
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48853 3c298f89-4303-0410-b956-a3cf2f4a3e73
According to the calling convention of the o32 ABI the caller
function must reserve stack space for $a0-$a3 registers in case
the callee needs to save its arguments.
The assembly code of the loader does not reserve stack space for
these registers thus when the 'loader_main' function needs to save
its arguments, those will be stored in the 'workspace' area instead
of the stack.
Because the workspace area is also used by other part of the code, the
saved register values gets overwritten and this often leads to failed
kernel boots.
Fix the code to reserve stack space for the registers to avoid this
error.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Backport of r48279
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48852 3c298f89-4303-0410-b956-a3cf2f4a3e73
The ubdev01 profile defines its own MTDPARTS with smaller firmware
partition, so give it its own UBNT_BOARD in mkfwimage.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r48828
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48847 3c298f89-4303-0410-b956-a3cf2f4a3e73
The hwid check was wrong, causing the AR71XX_MODEL value to end with a
space (as $hwver was unset).
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r47553
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48844 3c298f89-4303-0410-b956-a3cf2f4a3e73
The hardware of the v3 is identical to the v2.
Based-on-patch-by: Andreas Ziegler <github@andreas-ziegler.de>
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r47406
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48843 3c298f89-4303-0410-b956-a3cf2f4a3e73
Install uboot files to KERNEL_BUILD_DIR instead of BIN_DIR to fix
ImageBuilder. Similar fixes are necessary for many (all?) other uboot
targets.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r46605
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48835 3c298f89-4303-0410-b956-a3cf2f4a3e73
The DTS files aren't useful (other targets don't copy them either) and
clutter BIN_DIR with files unrelated to the chosen profile.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r46604
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48834 3c298f89-4303-0410-b956-a3cf2f4a3e73
This device is identical to the TL-WR740N v5, it even uses the same HWID (which
wasn't the case for older TL-WR741ND revisions).
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r46238
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48833 3c298f89-4303-0410-b956-a3cf2f4a3e73
TP-Link ships wr740n v5.0 to Romania (so probably everywhere except
China) with 4 MB flash and 32 MB memory (confirmed by their local
support). This patch adds support for the v5.0 clone of v4, i just
tested it on my own v5.0 router and it works.
Signed-off-by: Daniel Petre <daniel.petre@posteo.net>
Backport of r46115
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48832 3c298f89-4303-0410-b956-a3cf2f4a3e73
compat-wireless/backports now contains a bcm47xx_nvram.h file to
backport some of the functions in it which are used by the bcmfmac
driver. This file just checks for the kernel versions and provide an
empty implementations on older kernel versions. This is OK on most
systems, but on bcm47xx / bcm53xx systems we want to call the real
functions here. This commit removes the file from backports in our
build process like we do it with the bcma and ssb header files. Instead
we add a recent version into our kernel so all code uses only one
header file. On bcm47xx / bcm53xx the real implementations of this code
will be used.
Reported-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
[Backport of r47467. The recent mac80211 backport was missing this patch,
breaking the build of the brcmfmac module]
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48831 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48753
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48815 3c298f89-4303-0410-b956-a3cf2f4a3e73
This patch adds support for the Netgear EX2700 in Chaos Calmer and
builds approriate -sysupgrade and -factory images.
Signed-off-by: Joseph C. Lehner <joseph.c.lehner@gmail.com>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48792 3c298f89-4303-0410-b956-a3cf2f4a3e73
Backport patch for implementing sysupgrade support on brcm2708 target was
incomplete, since it was missing the default packages (r47604).
Add those packages in order to fully support sysupgrade.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48791 3c298f89-4303-0410-b956-a3cf2f4a3e73
This simply replaces init fix with a final version and puts it in a
generic dir. This will allow backporting some trivial changes from 4.6.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48788 3c298f89-4303-0410-b956-a3cf2f4a3e73
the u2_phy init was missing
Backport of r48747
Signed-off-by: John Crispin <blogic@openwrt.org>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48749 3c298f89-4303-0410-b956-a3cf2f4a3e73
Is it possible to backport commit r46269 (3c298f89-4303-0410-b956-a3cf2f4a3e73) to CC 15.05 in order to provide harddisk support.
Backporting tested (info available at: https://forum.openwrt.org/viewtopic.php?id=62389).
Thnx in advance!
Here is the original patch from trunk:
>From 76bfcd46bd55263947461e40bd02047922755c72 Mon Sep 17 00:00:00 2001
From: blogic <blogic@3c298f89-4303-0410-b956-a3cf2f4a3e73>
Date: Wed, 8 Jul 2015 14:25:36 +0000
Subject: [PATCH] malta: add harddisk support
works with:
qemu-system-mipsel -M malta \
-hda openwrt-malta-le-root.ext4 \
-kernel openwrt-malta-le-vmlinux.elf \
-nographic -append "root=/dev/sda console=ttyS0"
Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48666 3c298f89-4303-0410-b956-a3cf2f4a3e73
The image name for the HiLink HLK-RM04 module has a typo and should read "RM04" rather than "RM02"
Backport of r48355
Signed-off-by: John Clark <inindev@gmail.com>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48430 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48409
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48411 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48371
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48392 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48372
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48373 3c298f89-4303-0410-b956-a3cf2f4a3e73
Fully reset the chip like on a full up/down, but without the PHY
statemachine restart.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48228
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48246 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48227
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48245 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r47284
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48165 3c298f89-4303-0410-b956-a3cf2f4a3e73
It was never tested, most likely not working (because of le32_to_cpu)
and not upstreamed.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r48060
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48139 3c298f89-4303-0410-b956-a3cf2f4a3e73
Port 4 is the only ethernet port on this router, so disable all other PHYs
in order to save power.
Signed-off-by: Vittorio Gambaletta <openwrt@vittgam.net>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48097 3c298f89-4303-0410-b956-a3cf2f4a3e73
This patch allows configuring ports to be disabled in the device tree; this
saves power, since disabling ports here actually disables power to ethernet
PHYs.
Line 461 enables all ethernet ports, so line 508 is getting zero ports to be
disabled, except for port 5 in SoCs where this is not implemented as it will
be sticky disabled in register POC0. Because of this, the code will still read
the switch configuration and OR it to the device tree setting.
Signed-off-by: Vittorio Gambaletta <openwrt@vittgam.net>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48096 3c298f89-4303-0410-b956-a3cf2f4a3e73
Line 461 is actually enabling all switch ports by setting the disable bits
to 0. This needs to be done because the bootloader sets all ports to disabled
by default (which is the case for at least one router based on RT5350).
So, this patch fixes the comment in line 460.
Signed-off-by: Vittorio Gambaletta <openwrt@vittgam.net>
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48095 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48005
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48093 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r48085
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48092 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r47873
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48091 3c298f89-4303-0410-b956-a3cf2f4a3e73
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r47871
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48090 3c298f89-4303-0410-b956-a3cf2f4a3e73
My last bugfix (r47538) introduced a new bug in wndr3700_board_detect
(again...).
Assigning the result of ar71xx_get_mtd_offset_size_format to the model
variable before stripping of garbage using awk will cause all NUL bytes to
be removed before awk is applied, leading to model strings like
"NETGEAR WNDRMACv2NETGEAR", where a NUL byte after the v2 is supposed to
terminate the string.
Fix by calling ar71xx_get_mtd_offset_size_format twice, once piping to awk
directly.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r48075
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48089 3c298f89-4303-0410-b956-a3cf2f4a3e73
When fixing the model string for WNDR3700v2 which contain a model string
followed by 0xff in r46455, the match for other versions of the WNDR3700v2
which just contain lots of 0xff broke (as the 0xff $model is checked for
is stripped off).
Fix by stripping off non-printable characters only for the actual output
string, but not for the internal matching.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r47538
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48088 3c298f89-4303-0410-b956-a3cf2f4a3e73
There are a few NETGEAR devices which don't terminate the model name in the
ART with a NUL byte, at least some NETGEAR WNDR3700v2. The current awk
expression doesn't match 0xFF bytes, so AR71XX_MODEL contains lots of
trailing 0xFF garbage in this case.
Fix this by matching for the first non-printable character and explicitly
setting LC_CTYPE=C (probably not strictly necessary on OpenWrt, but will
definitely work like this, even when awk supports locales and LANG is set).
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r46455
git-svn-id: svn://svn.openwrt.org/openwrt/branches/chaos_calmer@48087 3c298f89-4303-0410-b956-a3cf2f4a3e73
