While an image layout based on MBR and 'bootfs' partition may be easy
to understand for users who are very used to the IBM PC and always have
the option to access the SD card outside of the device (and hence don't
really depend on other recovery methods or dual-boot), in my opinion
it's a dead end for many desirable features on embedded systems,
especially when managed remotely (and hence without an easy option to
access the SD card using another device in case things go wrong, for
example).
Let me explain:
* using a MSDOS/VFAT filesystem to store kernel(s) is problematic, as a
single corruption of the bootfs can render the system into a state
that it no longer boots at all. This makes dual-boot useless, or at
least very tedious to setup with then 2 independent boot partitions
to avoid the single point of failure on a "hot" block (the FAT index
of the boot partition, written every time a file is changed in
bootfs). And well: most targets even store the bootloader environment
in a file in that very same FAT filesystem, hence it cannot be used
to script a reliable dual-boot method (as loading the environment
itself will already fail if the filesystem is corrupted).
* loading the kernel uImage from bootfs and using rootfs inside an
additional partition means the bootloader can only validate the
kernel -- if rootfs is broken or corrupted, this can lead to a reboot
loop, which is often a quite costly thing to happen in terms of
hardware lifetime.
* imitating MBR-boot behavior with a FAT-formatted bootfs partition
(like IBM PC in the 80s and 90s) is just one of many choices on
embedded targets. There are much better options with modern U-Boot
(which is what we use and build from source for all targets booting
off SD cards), see examples in mediatek/mt7622 and mediatek/mt7623.
Hence rename the 'sdcard' feature to 'legacy-sdcard', and prefix
functions with 'legacy_sdcard_' instead of 'sdcard_'.
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
procd.sh:
Instead of triggering on every mount.add event, there should be no
mount trigger at all in case none of the directories passed to
procd_add_*_mount_trigger() are located on a mountpoint configured in
/etc/config/fstab.
uxc:
add missing dependency on rpcd.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
040fecc system: fix issues reported by Coverity
48f481b service: make sure string read is null terminated
16dbc2a uxc: fix a bunch of issues discovered by Coverity
ff9002f uxc: fix help output
104b49d uxc: support config in uvol
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
This resolves incosnsitencies of the configured RX / TX flow control
modes between different boards or bootloaders.
Signed-off-by: David Bauer <mail@david-bauer.net>
The chip supports clock speeds up to 50 MHz, however it won't even read
the chip-id correctly at this frequency.
45 MHz however works reliable.
Signed-off-by: David Bauer <mail@david-bauer.net>
Delete tunnel on 6in4 interface teardown.
Should solve problem related to tunnel stuck on restart loop
with "Unknown Command" on tunnel restart due to wan connection drop.
Fixes: FS#3690
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
When a target configuration has unser Kconfig symbols, the build will
fail when OpenWrt is compiled with V=s and stdin is connected to a tty.
In case OpenWrt is compiled without either of these preconditions, the
build will uscceed with the symbols in question being unset.
Modify the kernel configuration in a way it fails on unset symbols
regardless of the aformentioned preconditions.
Signed-off-by: David Bauer <mail@david-bauer.net>
Calling free for the OF property can result in a kernel panic, as the
buffer in question might be referenced elsewhere. Also, it is not
removed from the tree.
Always allocate a new property and updating the tree with it fixes both
issues.
Fixes commit 91a52f22a1 ("treewide: backport support for nvmem on non platform devices")
Signed-off-by: David Bauer <mail@david-bauer.net>
This was missed because scancode license scanner was confused by
comments about crc32buf().
Cc: Gabor Juhos <juhosg@openwrt.org>
Cc: Gabor Juhos <juhosg@freemail.hu>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This was missed because scancode license scanner was confused by a
comment about (no) copyrights in the init_crc_table().
Cc: Gabor Juhos <juhosg@openwrt.org>
Cc: Gabor Juhos <juhosg@freemail.hu>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This was missed because scancode license scanner was confused by a
comment about Cisco's GPL code github repository.
Cc: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Acked-by: Christian Lamparter <chunkeey@gmail.com>
The EXT4 driver also takes care of EXT2 and EXT3 file systems.
Activating the EXT2 driver kernel config options unlocked some other
ext2 driver related options which OpenWrt did not take care of.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The change which backported the of_get_mac_address() change broke some
patches in the layerscape target so the patches did not apply any more.
This commit makes them apply again and also fixes some other problems
related to this change.
Fixes commit 91a52f22a1 ("treewide: backport support for nvmem on non platform devices")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The code from ks8851.c was moved to ks8851_common.c, so it was not
backported. This broke the compile of the omap target which uses this
driver.
Fixes commit 91a52f22a1 ("treewide: backport support for nvmem on non platform devices")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This patch is backported from linux-arm-kernel [1] to improve situation, when
it was reported that 1.2 GHz variant is unstable with DFS.
It waits to be accepted upstream, however, it waits for Marvell people to respond.
[1] https://patchwork.kernel.org/project/linux-arm-kernel/patch/20210630225601.6372-1-kabel@kernel.org/
Fixes: 7b868fe04a ("Revert "mvebu: 5.4 fix DVFS caused random boot crashes"")
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Based on the discussion on the mailing list [1], the patch which was
reverted, it reverts only one patch without the subsequent ones.
This leads to the SoC scaling issue not using a CPU parent clock, but
it uses DDR clock. This is done for all variants, and it's wrong because
commits (hacks) that were using the DDR clock are no longer in the mainline kernel.
If someone has stability issues on 1.2 GHz, it should not affect all
routers (1 GHz, 800 MHz) and it should be rather consulted with guys, who are trying to
improve the situation in the kernel and not making the situation worse.
There are two solutions in cases of instability:
a) disable cpufreq
b) underclock it up to 1 GHz
This reverts commit 080a0b74e3.
[1] https://lists.openwrt.org/pipermail/openwrt-devel/2021-June/035702.html
Fixes: d379476817 ("mvebu: armada-37xx: add patch to forbid cpufreq for 1.2 GHz")
CC: Pali Rohár <pali@kernel.org>
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Add --disable-libdebuginfod with remove libcurl dependency.
Remove totally unused host elfutils.
Refreshed and rebased patches.
Also happens to fix compilation with GCC11.
Newer versions of elfutils seem to have some kind of dependency on
obstack.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
When using htmode 'HE20' with a radio mode that uses wpa-supplicant
(like mesh or sta), it will default to 40 MHz bw if disable_ht40 is not
set. This commit fixes this behaviour.
Signed-off-by: Jesus Fernandez Manzano <jesus.manzano@galgus.net>
This enables building of rpcapd and adds it as a package.
It is a daemon that allows remote packet capturing from another machine.
E.g. Wireshark can talk to it using the Remote Capture Protocol (RPCAP).
https://www.tcpdump.org/manpages/rpcapd.8.html
Compile and run tested: OpenWrt SNAPSHOT r17190-2801fe6132 on x86/64
Signed-off-by: Stephan Schmidtmer <hurz@gmx.org>
When having two keys that start with the same characters and the second
key just has one character more nand_tffs_read and tffs_read return the
wrong value for the longer key. This is due to the usage of strncmp in
combination with the length of the shorter key which is usually first in
the list before the longer key and when strncmp matches, the search is
stopped. The problem only occurs when the length of the two keys is
different, not if just the last character is different. The fix is to
use strcmp and as such it will only return the value if the key (name)
and the key to look for (namefilter) have the same value and length. A
sample case returning wrong values is when keys macwlan and macwlan2 are
defined and querying macwlan2 returns the value for macwlan.
Signed-off-by: Daniel Kestrel <kestrel1974@t-online.de>
