The skb->len field is read after the packet is sent to the network
stack. In the meantime, skb can be freed. This patch fixes this bug.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
After enhancing ltq-deu, build it by default for the devices
using it.
Reverts: 964863b ("ltq-deu: Mark lantiq DEU broken")
Signed-off-by: Daniel Kestrel <kestrel1974@t-online.de>
This patch includes a series of performance improvements. All patches
were accepted and should land in 5.17.
NAT Performance results on BT Home Hub 5A (kernel 5.10.89, mtu 1500):
Down Up
Before 539 Mbps 599 Mbps
After 624 Mbps 695 Mbps
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
xrx200 max MTU is reduced so that it works correctly when set to the
max, and the max MTU of the switch is increased to match.
In 5.10, the switch driver now enables non-standard MTUs on a per-port
basis, with the overall frame size set based on the cpu port. When the
MTU is not used, this should have no effect. The maximum packet size is
limited as large packets cause the switch to lock up.
0702-net-lantiq-add-support-for-jumbo-frames.patch comes from net-next
commit 998ac358019e491217e752bc6dcbb3afb2a6fa3e.
In 5.4, all switch ports are configured to accept the max MTU, as 5.4
does not have port_max_mtu/port_change_mtu callbacks.
Signed-off-by: Thomas Nixon <tom@tomn.co.uk>
It was reported the caldata on the FritzBox 7430 is not only stored at
different offsets, but is also larger than the current output size
limit.
Increase the output file size limit (after deflate) by 1024 bytes.
Ref: FS#3604 ("ath9k firmware is 0 bytes on Fritzbox 7430")
Signed-off-by: David Bauer <mail@david-bauer.net>
It was reported the AVM FritzBox 7430 has different offsets for the
caldata depending on the device.
Whether this is due to custom bad-block handling or up to the installed
bootloader-version is unknown.
Try both known caldata offsets like it is already done for the ipq40xx
NAND based models. Use the same approach for the FritzBox 7412. While it
is currently unknown, whether it is affected, applying the same logic
has no downsides.
Signed-off-by: David Bauer <mail@david-bauer.net>
All buttons of the FritzBox 7360 family are active-low, not active-high.
Corrent the GPIO flag. This fixes release triggers upon push of a button.
Reported-by: Jan-Niklas Burfeind <git@aiyionpri.me>
Signed-off-by: David Bauer <mail@david-bauer.net>
Each of
- CRYPTO_AEAD2
- CRYPTO_AEAD
- CRYPTO_GF128MUL
- CRYPTO_GHASH
- CRYPTO_HASH2
- CRYPTO_HASH
- CRYPTO_MANAGER2
- CRYPTO_MANAGER
- CRYPTO_NULL2
either directly required for mac80211 crypto support, or directly
selected by such options. Support for the mac80211 crypto was enabled in
the generic config since c7182123b9 ("kernel: make cryptoapi support
needed by mac80211 built-in"). So move the above options from the target
configs to the generic config to make it clear why do we need them.
CC: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
Both CLANG_VERSION and LLD_VERISON are autogenerated runtime
configuration options, so add them to the kernel configuration filter
and remove from generic and per-target configs to keep configs clean.
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
adds the correct offset for the calibration data.
The values are according to the OpenWrt Forum Thread:
"Fritzbox 7430 and wifi".
Link: <https://forum.openwrt.org/t/fritzbox-7430-and-wifi/86944>
Reported-by: RENErica
Signed-off-by: Joel Linn <jl@conductive.de>
[changed commit message]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
The xway_legacy subtarget only supports 5 devices. Most compiled
switch drivers are unused by any of these devices. The same drivers
are compiled into the xway subtarget. They were probably copied
from there when creating this subtarget.
Switches used by devices:
Arcadyan ARV4518PWR01 Realtek RTL8306SD
Arcadyan ARV4518PWR01A Realtek RTL8306SD
Arcadyan ARV4520PW Infineon ADM6996I
Arcadyan ARV4525PW only PHY(IC+ IP101A)
Arcadyan ARV452CQW Realtek RTL8306
The CONFIG_ETHERNET_PACKET_MANGLE symbol has also been disabled,
as it is only needed by the driver for AR8216.
Reduces kernel size by 19.9 kB.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
The board has a fixed size kernel partition but do not limit the kernel
size during image building.
Disable image building for both boards as well, since the kernel of the
last release as well as master are to big to fit into the 2 MByte kernel
partition.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The current state of the kernel 5.4 support is in the openwrt-21.02
branch. No need to keep a not default used kernel in this branch.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de> [VRX268/ bthub5]
With Kernel 5.10 the ar7 FRITZ!Box are not booting the initramfs nor the
sysupgrade image any more. Presumably due to the grown kernel.
Use the okli preloader to workaround the bootloader issue. No solution
so far for the initramfs.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Removed due to being unused with 1f7a03a706, but now required for the
ar7 FRITZ!Box.
Could be used for the ARV7519RW22 as well, for which the image
generation was disabled due to a stock u-boot issue with kernel bigger
than 2 MByte.
The code is combination of the ath79 and ramips okli loader.
Signed-off-by: Mathias Kresin <dev@kresin.me>
This commit contains a series of fixes for DMA. The burst length
patch significantly improves Ethernet performance. Patches were
tested on the xRX200 and xRX330.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
The devicetree property mac-address is expected to be set by the
bootloader and has priority over the nvmem supplied one.
Drop the mac-address address property from the dtsi files, to let the
mac address from nvmem-cells get used.
Signed-off-by: Mathias Kresin <dev@kresin.me>
OpenWrt maintains two special out-of-tree DT properties:
"qca,disable-5ghz" and "qca,disable-2ghz". These are implemented
in a mac80211 ath9k patch "550-ath9k-disable-bands-via-dt.patch".
With the things being what they are, now might be a good
point to switch the devices to the generic and upstream
"ieee80211-freq-limit" property. This property is much
broader and works differently. Instead of disabling the
drivers logic which would add the affected band and
channels. It now disables all channels which are not
within the specified frequency range.
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Tested-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com> # HH5A
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
When trying to add support for another device with Micron NAND chips,
it was discovered that the default setting in the kernel source does
not work with Micron Chips, since the device trees setting is
overwritten and hard coded by the kernel xway_nand driver. This was
the original reason for this PR.
A kernel patch sets the default ECC mode to soft without overwriting
the device tree settings and the device tree for devices using it
are updated with new parameters because the old ones are deprecated
by torvalds/linux@533af69.
A patch for kernel 5.4 is provided to support the new settings
because kernel 5.4 does not support it.
Signed-off-by: Daniel Kestrel <kestrel1974@t-online.de>
This backports a fix proposed for upstream kernel to fix overwriting
the NAND ECC engine in device tree.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
CONFIG_RCU_{NEED_SEGCBLIST,STALL_COMMON} are set basically everywhere. Move them
to the generic kconfigs. And resort the generic kconfigs while at it.
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
In the current state, nvmem cells are only detected on platform device.
To quickly fix the problem, we register the affected problematic driver
with the of_platform but that is more an hack than a real solution.
Backport from net-next the required patch so that nvmem can work also
with non-platform devices and rework our current patch.
Drop the mediatek and dsa workaround and rework the ath10k patches.
Rework every driver that use the of_get_mac_address api.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
Define nvmem-cells and convert mtd-mac-address to nvmem implementation.
The conversion is done with an automated script.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
Rework patch 681-NET-add-mtd-mac-address-support to implement
only the function to read the mac-address from mtd.
Generalize mtd-mac-address-increment function so it can be applied
to any source of of_get_mac_address.
Rename any mtd-mac-address-increment to mac-address-increment.
Rename any mtd-mac-address-increment-byte to mac-address-increment-byte.
This should make simplify the conversion of target to nvmem implementation.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
sysupgrade metadata is not flashed to the device, so check-size
should be called _before_ adding metadata to the image.
While at it, do some obvious wrapping improvements.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Acked-by: Paul Spooren <mail@aparcar.org>
When the ltq_deu_vr9 kernel module is loaded, hostapd does not start any
more. It fails with this error message:
daemon.err hostapd: nl80211: kernel reports: key addition failed
daemon.err hostapd: Interface initialization failed
OpenWrt uses the standard Linux crypto API in the wifi drivers now
and this probably makes the system offload more crypto operations to
special hardware like the Lantiq DEU. There is probably a bug in the DEU
and these operations fail and then hostapd does not start the interface.
Do not include the Lantiq DEU by default any more.
Fixes: FS#3901
Fixes: 53b6783907 ("mac80211: remove patches stripping down crypto support")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Acked-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Tested-by: Notupus <notpp46@gmail.com>
Enable the XRX200 PMAC, GSWIP DSA tag and GSIP DSA drivers in the 5.4
kernel config. Update the existing vr9_*.dts{,i} to use the new
Ethernet and switch drivers. Drop the swconfig package from the xrx200
target because swconfig doesn't manage DSA based switches.
The new /etc/config/network format for the DSA driver is not compatible
with the old (swconfig) based one. Show a message during sysupgrade
notifying users about this change and asking them to start with a fresh
config (or forcefully update and then migrate the config manually).
Failsafe mode can now automatically bring up the first lan interface
based on board.json including DSA based setups. Drop
05_set_preinit_iface_lantiq from the xRX200 sub-target as this is not
needed anymore. For now we are keeping it for the ase, xway and
xway_legacy until there's some confirmation that it can be dropped from
there as well.
While here, some boards also receive minor fixups:
- Use LAN1 as LAN1 (according to a photo this port can also be
configured as WAN) on the Buffalo WBMR-300HPD. This makes it easier to
read the port mapping because otherwise we would have LAN{2,3,4} and
WAN (which was the case for the non-DSA version previously).
- vr9_avm_fritz3390.dts: move the "gpio" comment from port 0 and 1 to
their corresponding PHYs
- vr9_tplink_vr200.dtsi: move the "gpio" comment from port 0 to PHY 0
- vr9_tplink_tdw89x0.dtsi: move the "gpio" comment from port 0 to PHY 0
Acked-by: Aleksander Jan Bajkowski <A.Bajkowski@stud.elka.pw.edu.pl>
Tested-by: Notupus <notpp46@googlemail.com> # TD-W9980/DM200/FRITZ 7430
Tested-by: Martin Schiller <ms@dev.tdt.de> # tested on TDT VR2020
Tested-by: Martin Schiller <ms@dev.tdt.de> # tested on TP-Link TD-W8980B
Tested-by: Martin Schiller <ms@dev.tdt.de> # tested on ZyXEL P-2812HNU-F1
Tested-by: Daniel Kestrel <kestrel1974@t-online.de> # tested on Fritzbox 7490
Tested-by: Daniel Kestrel <kestrel1974@t-online.de> # tested on Fritzbox 3490
Tested-by: @jospezial <jospezial@gmx.de> # tested on VGV7510KW22 (o2 Box 6431)
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Many people appear to use an unneeded "+" prefix for the increment
when calculating a MAC address with macaddr_add. Since this is not
required and used inconsistently [*], just remove it.
[*] As a funny side-fact, copy-pasting has led to almost all
hotplug.d files using the "+", while nearly all of the
02_network files are not using it.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Removed upstreamed:
generic/backport-5.4/050-gro-fix-napi_gro_frags-Fast-GRO-breakage-due-to-IP-a.patch
bcm63xx/patches-5.4/434-nand-brcmnand-fix-OOB-R-W-with-Hamming-ECC.patch*
Removed/code was included upstream and therefore redundant:
ramips/patches-5.4/999-fix-pci-init-mt7620.patch
All other patches automatically rebased.
* update_kernel.sh did not flag this yet it was included in 5.4.119[1], as a
result of the rebase, I removed my testing lines since I did not go back to
test built or to run test 5.4.119 with the removed patch present.
1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.4.119&id=e5b3e69eb36ac1178a7a2392616fd29afd288c4e
Signed-off-by: John Audia <graysky@archlinux.us>
ath79, lantiq, ipq40xx, ramips all use the OpenWrt-specific gpio-export
functionality. Consolidate the patch that adds it under hack-5.10 since
this logic is obviously not target-specific. For those who want to
disable it, unsetting CONFIG_GPIO_SYSFS symbol will disable this code.
Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>
CONFIG_FORTIFY_SOURCE=y is already set in the generic kernel
configuration, but it is not working for MIPS on kernel 5.4, support for
MIPS was only added with kernel 5.5, other architectures like aarch64
support FORTIFY_SOURCE already since some time.
This patch adds support for FORTIFY_SOURCE to MIPS with kernel 5.4,
kernel 5.10 already supports this and needs no changes.
This backports one patch from kernel 5.5 and one fix from 5.8 to make
fortify source also work on our kernel 5.4.
The changes are not compatible with the
306-mips_mem_functions_performance.patch patch which was also removed
with kernel 5.10, probably because of the same problems. I think it is
not needed anyway as the compiler should automatically optimize the
calls to memset(), memcpy() and memmove() even when not explicitly
telling the compiler to use the build in variant.
This increases the size of an uncompressed kernel by less than 1 KB.
Acked-by: Rosen Penev <rosenp@gmail.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Manually rebased*
generic/backport-5.4/700-v5.5-net-core-allow-fast-GRO-for-skbs-with-Ethernet-heade.patch
Added new backport*
generic/backport-5.4/050-gro-fix-napi_gro_frags-Fast-GRO-breakage-due-to-IP-a.patch
All others updated automatically.
The new backport was included based on this[1] upstream commit that will be
mainlined soon. This change is needed because Eric Dumazet's check for
NET_IP_ALIGN (landed in 5.4.114) causes huge slowdowns on drivers which use
napi_gro_frags().
Build system: x86_64
Build-tested: ipq806x/R7800
Run-tested: ipq806x/R7800
No dmesg regressions, everything functional
*Credit to Alexander Lobakin
1. https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/commit/?id=7ad18ff6449cbd6beb26b53128ddf56d2685aa93
Signed-off-by: John Audia <graysky@archlinux.us>
