This backports some patches from kernel 5.15 to fix issues with
flowtable offloading in kernel 5.10. OpenWrt backports most of the
patches related to flowtable offloading from kernel 5.15 already, but we
are missing some of the extra fixes.
This fixes some connection tracking problems when a flow gets removed
from the offload and added to the normal SW path again.
The patch 614-v5.18-netfilter-flowtable-fix-TCP-flow-teardown.patch was
extended manually with the nf_conntrack_tcp_established() function.
All changes are already included in kernel 5.15.
Fixes: #8776
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 96ef2dabce)
The nft NAT packages for IPv4 and IPv6 were merged into the common
packages with kernel 5.1. The kmod-nft-nat6 package was empty in our
build, remove it.
Multiple kernel configuration options were also removed, remove them
from our generic kernel configuration too.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit b75425370d)
Use the kernel's built-in formula for computing this value.
The value applied by OpenWRT's sysctl configuration file does not scale
with the available memory, under-using hardware capabilities.
Also, that formula also influences net.netfilter.nf_conntrack_buckets,
which should improve conntrack performance in average (fewer connections
per hashtable bucket).
Backport upstream commit for its effect on the number of connections per
hashtable bucket.
Apply a hack patch to set the RAM size divisor to a more reasonable value (2048,
down from 16384) for our use case, a typical router handling several thousands
of connections.
Signed-off-by: Vincent Pelletier <plr.vincent@gmail.com>
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
(cherry picked from commit 15fbb91666)
|Warning: trailing whitespace in line 66 of drivers/mtd/parsers/Kconfig
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit d6801e0d3f)
All patches automatically rebased.
The following patch was replaced by a similar version upstream:
bcm27xx/patches-5.10/950-0036-tty-amba-pl011-Add-un-throttle-support.patch
Signed-off-by: John Audia <therealgraysky@proton.me>
(cherry picked from commit 7d3c0928de)
1. KCFLAGS should be used for custom flags
2. Optimization flags are arch / SoC specific
3. -fno-reorder-blocks may *worsen* network performace on some SoCs
4. Usage of flags was *reversed* since 5.4 and noone reported that
If we really need custom flags then CONFIG_KERNEL_CFLAGS should get
default value adjusted properly (per target).
Ref: 4e0c54bc5b ("kernel: add support for kernel 5.4")
Link: http://lists.openwrt.org/pipermail/openwrt-devel/2022-June/038853.html
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20190409093046.13401-1-zajec5@gmail.com/
Cc: Felix Fietkau <nbd@nbd.name>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Cc: Rui Salvaterra <rsalvaterra@gmail.com>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 22168ae681)
This uses kernel's generic variable and doesn't require patching it with
a custom Makefile change. It's expected *not* to change any behaviour.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 1d42af720c)
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 24e27bec9a)
When building the mediatek/mt7629 target in OpenWrt 22.03 the kernel
does not have a configuration option for CONFIG_CRYPTO_DEV_MEDIATEK. Add
this option to the generic kernel configuration and also add two other
configuration options which are removed when we refresh the mt7629
kernel configuration.
Fixes: 2bea35cb55 ("mediatek: remove crypto-hw-mtk package")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit dcc0fe24ea)
Update this pending patch to remove the untested (variable eraseregions)
section, alongside simplifying the patch.
Signed-off-by: John Thomson <git@johnthomson.fastmail.com.au>
[refresh and split out unrelated refreshes]
Signed-off-by: Thibaut VARÈNE <hacks@slashdirt.org>
(cherry picked from commit 4f7065ed25)
Kernel 5.10.124 introduced a new symbol 'LIB_MEMNEQ'. Add it to the
generic 5.10 config.
Fixes: 9e5d743422 ("kernel: bump 5.10 to 5.10.124")
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
(cherry picked from commit f3caba679b)
This is now built-in, enable so it won't propagate on target configs.
Link: https://lkml.org/lkml/2022/1/3/168
Fixes: 79e7a2552e ("kernel: bump 5.15 to 5.15.44")
Fixes: 0ca9367069 ("kernel: bump 5.10 to 5.10.119")
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
(Link to Kernel's commit taht made it built-in,
CRYPTO_LIB_BLAKE2S[_ARM|_X86] as it's selectable, 5.10 backport)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 539e60539a)
Hannu Nyman wrote in openwrt's github issue #9962:
|Based on forum discussion, the commit 0bc794a
|"kernel: add support for Toshiba TC58NVG0S3HTA00 NAND flash"
|causes flash memory chip misdetection for some other
|Fritzbox devices, as the commit only defines a 4-byte flash
|memory chip ID that matches several chips used in the devices.
|
|See discussion from this onward
|<https://forum.openwrt.org/t/openwrt-22-03-0-rc1-first-release-candidate/126045/182>
|
|OpenWrt 22.03.0-rc2 and rc3 are causing on a Fritzbox 7412
|bootloops due to a misdetected flash chip.
|
|Yup, that patch is missing the 5th ID byte entirely - both chips
|share the same first 4;
|
| TC58NVG0S3HTA00 = 0x98 0xf1 0x80 0x15 0x72 (digikey datasheet, page 35)
| TC58BVG0S3HTA00 = 0x98 0xf1 0x80 0x15 0xf2 (digikey datasheet, page 28)
|
|The commit has also been backported to openwrt-22.03 after rc1,
|so both rc2 and rc3 suffer from this bug."
Andreas' TC58NVG0S3H seems not to follow Toshibas/Kioxa's own datasheet.
It only reports the first four bytes: "98 f1 80 15 00 00 00 00".
This patch changes the id_len in the entry to 8. This makes it so that
Andreas' NAND is still detected. At the same time, this prevents other
Toshiba NAND flash chips - that share the same four bytes - from being
misdetected.
Upstream (Miquel Raynal) decided to drop this patch for now. But he
advised to keep it in OpenWrt. As other devices could be affected.
<https://lore.kernel.org/linux-mtd/20220606155919.23001410@xps-13/>
Reported-by: Peter-vdL
Tested-by: Peter-vdL
Tested-by: Andreas B<C3><B6>hler <dev@aboehler.at>
Fixes: 0bc794a668 ("kernel: add support for Toshiba TC58NVG0S3HTA00 NAND flash")
Link: <https://github.com/openwrt/openwrt/issues/9962>
(actually move the patch, added comment about possible counterfeits)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Delete the crypto-lib-blake2s kmod package, as BLAKE2s is now built-in.
Patches automatically rebased.
Build system: x86_64
Build-tested: ipq806x/R7800, x86/64
Signed-off-by: John Audia <therealgraysky@proton.me>
(cherry picked from commit cd634afe6c)
- fix an issues when accessing the port pointer of an expired/invalid fdb entry
Signed-off-by: Felix Fietkau <nbd@nbd.name>
(cherry picked from commit 38a5b593ef)
Some dst in IPv6 flow offload table become invalid after the table is created.
So check_dst is needed in packet path.
Signed-off-by: Ritaro Takenaka <ritarot634@gmail.com>
[Add patch for kernel 5.15 too and rename file]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit efff48529b)
This backports a patch from Linux 5.10.116 to fix a compile problem
introduced in 5.10.114.
drivers/usb/phy/phy-generic.c could not find
devm_regulator_get_exclusive().
Fixes: 8592df67f4 ("kernel: bump 5.10 to 5.10.114")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 7400adae8d)
The Toshiba TC58NVG0S3HTA00 is detected with 64 byte OOB while the flash
has 128 byte OOB. This adds a static NAND ID entry to correct this.
Signed-off-by: Andreas Böhler <dev@aboehler.at>
(cherry picked from commit 0bc794a668)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
This series also contains other improvement for hardware flow offload support
Signed-off-by: Felix Fietkau <nbd@nbd.name>
(cherry-picked from commit 0f029b3d2b)
MPLS feature symbols are normally only set when kmod-mpls is enabled, but the
CONFIG_MPLS symbol they depend on could also have been selected by openvswitch
instead
Signed-off-by: Felix Fietkau <nbd@nbd.name>
(cherry-picked from commit 92add80414)
There are various reports on Github and in the forum that this commit
causes multiple problems.
This reverts commit ee6ba216d8.
Fixes: #9420
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
(cherry picked from commit 3e0daca644)
This patch was accidentally backported from master for kernel 5.15 too.
Remove the version for kernel 5.15 and keep the version for kernel 5.10.
Fixes: 9ab337dfbc ("kernel: backport pgalloc memory leak fix")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This is required to support built-in modem of ZTE MF286R, in addition to
other external modems, such as MF831, MF910, MF920, which refuse to
reconfigure their remote MAC address, even if "locally administered" bit
is set, leading to dropped traffic towards the host. Add a workaround
for that issue already present in cdc_ether to rndis_host driver as
well.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
(cherry picked from commit c99013e242)
The correct readid method is SPINAND_READID_METHOD_OPCODE_ADDR.
Signed-off-by: Felix Matouschek <felix@matouschek.org>
(cherry picked from commit 3711aee56d)
# CONFIG_PSTORE_COMPRESS_DEFAULT="deflate"
this can lead to confusion. Thankfully, in the KConfig
world this setting is still interpreted as disabled.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit de4879c1ae)
Backport a fix for the massive memory leak observed in Octeon after
switching to kernel 5.10.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
(cherry picked from commit 9283359bd5)
In commit ab143647ef ("kernel: generic: improve FIT partition parser")
part_bits was bumped to 2 in order to allow up to 3 additional FIT
sub-images mapped into sub-partitions.
This change has to be reflected also in our local patch
420-mtd-set-rootfs-to-be-root-dev.patch
which still assumed part_bits==1 for mtdblock devices in case of
CONFIG_FIT_PARTITION=y.
Fixes: #9557
Fixes: ab143647ef ("kernel: generic: improve FIT partition parser")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit 13960fb0e0)
* only map filesystems configured in 'loadables'
* allow mapping more than one filesystem (e.g. customization/branding
or localization in addition to rootfs)
* small cleaning here and there
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit ab143647ef)
If the selected boot configuration is stored by U-Boot in '/chosen'
node as 'bootconf' attribute, use that configuration to resolve the
block device used as rootfs. Fall back to use the default configuration
in case 'bootconf' is not present.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit 503f3b9f0e)
at91/sama7 fails to build due to:
| Asymmetric (public-key cryptographic) key type (ASYMMETRIC_KEY_TYPE) [Y/?] y
| Asymmetric public-key crypto algorithm subtype (ASYMMETRIC_PUBLIC_KEY_SUBTYPE) [Y/?] y
| Asymmetric TPM backed private key subtype (ASYMMETRIC_TPM_KEY_SUBTYPE) [N/m/?] (NEW)
|Error in reading or end of file.
please note that asym_tpm (module) has been removed in 5.17:
<https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d3cff4a9>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 6387715093)
Remove "a" character from the first line of patch
738-v5.14-01-net-dsa-qca8k-fix-an-endian-bug-in-qca8k-get-ethtool.patch
Otherwise `git am` fails to apply this patch which is annoying when
trying to do some development / rebasing.
Signed-off-by: Marek Behún <kabel@kernel.org>
(cherry picked from commit f811c33b19)
As the upcoming release will be based on Linux 5.10 only, remove all
kernel configuration as well as patches for Linux 5.4.
There were no targets still actively using Linux 5.4.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit 3a14580411)
Kernel 5.6 introduced a new config symbol SERIAL_8250_16550A_VARIANTS.
In kernel 5.8, this symbol was changed to default to on on !x86, as some
embedded devices still use 16650A variants. Let's play safe here and
enable this symbol in the generic config, to avoid others from running
into this problem and having to spend several hours trying to bisect
this problem. While we could disable the symbol in the x86 target
configs, a 20ms boot time reduction really isn't worth the time wasted
on bisecting this issue.
Matt discovered this problem while working on adding support for the
WatchGuard Firebox M200 to the qoriq target, where it caused some
characters to be missing on the console output.
Reported-by: Matt Fawcett <mattytap@icloud.com>
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Reviewed-by: Rui Salvaterra <rsalvaterra@gmail.com>
Seeing failure to build because of missing symbols related to provisioning
CONFIG_KEXEC and signed images. Without this, if you set
CONFIG_KERNEL_KEXEC=y and try to build, target/linux will hang at:
scripts/kconfig/conf --syncconfig Kconfig
...
kexec system call (KEXEC) [Y/n/?] y
kexec file based system call (KEXEC_FILE) [Y/n/?] y
Verify kernel signature during kexec_file_load() syscall (KEXEC_SIG) [N/y/?] (NEW)
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Backport patches
381a730182f1 ("net: dsa: Move VLAN filtering syncing out of dsa_switch_bridge_leave")
108dc8741c20 ("net: dsa: Avoid cross-chip syncing of VLAN filtering")
from upstream (currently in net-next) to fix null-pointer dereference.
Signed-off-by: Marek Behún <kabel@kernel.org>
Add patch found in Teltonika RUT9_R_00.07.01.4 GPL SDK download[1]
adding USB IDs of the MeigLink SLM750 to the relevant kernel drivers.
Newer versions of Teltonika's 2G/3G/4G RUT9XX WWAN router series come
with this kind of modem.
[1]: https://wiki.teltonika-networks.com/view/GPL
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
This patch adds a patch to allow using parser_trx from ramips target,
mainly for Buffalo devices.
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
This patch moves the patches of parser_trx in mediatek target to
generic/backport-5.10 to use the changes from ramips target and
backport the additional patch of the parser.
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
The XMC XM25QH128C is a 16MB SPI NOR chip. The patch is verified on Ruijie RG-EW3200GX PRO.
Datasheet available at https://www.xmcwh.com/uploads/435/XM25QH128C.pdf
Signed-off-by: Langhua Ye <y1248289414@outlook.com>
the Aerohive HiveAP-330 and HiveAP-350 come equipped
with an TI TMP125 temperature chip. This patch wires
up the necessary support for this sensor and exposes
it through hwmon / thermal sensor framework. Upstream
support is coming, but it has to go through hwmon-next
first.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
The last remaining 5.4 target currently chokes because the
symbols haven't been disabled like for 5.10.
Fixes: 97158fe10e ("kernel: package ramoops pstore-ram crash log storage")
Reported-by: Hannu Nyman <hannu.nyman@iki.fi>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Add the following kconfig symbols (disabled):
CONFIG_DEFAULT_FQ
CONFIG_DEFAULT_CODEL
CONFIG_DEFAULT_SFQ
Also resort the config with the kconfig.pl script.
Fixes: f39872d966 ("kernel: generic: select the fq_codel qdisc by default")
Tested-by: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
Package the ability to log kernel crashes to 'ramoops' pstore
files into RAM in /sys/fs/pstore
Reference to the ramoops admin guide in upstream Linux:
https://www.kernel.org/doc/html/v5.10/admin-guide/ramoops.html
The files in RAM survive a warm reboot, but not a cold reboot.
Note: kmod-ramoops selects kmod-pstore and kmod-reed-solomon.
The feature can be used by selecting the kmod-ramoops and
adding a ramoops reserved-memory definition to the device DTS.
Example from R7800:
reserved-memory {
rsvd@5fe00000 {
reg = <0x5fe00000 0x200000>;
reusable;
};
ramoops@42100000 {
compatible = "ramoops";
reg = <0x42100000 0x40000>;
record-size = <0x4000>;
console-size = <0x4000>;
ftrace-size = <0x4000>;
pmsg-size = <0x4000>;
};
};
If no definition has been made in DTS, no crash log is stored
for the device.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(added CONFIG_EFI_VARS_PSTORE disable)
Add the missing IPv6 flow offloading support for routing only.
Hardware flow offloading is done by the packet processing engine (PPE)
of the Ethernet MAC and as it doesn't support mangling of IPv6 packets,
IPv6 NAT cannot be supported.
Signed-off-by: David Bentham <db260179@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
flowtable->net was initialized too late, and this could be triggered even
without hardware offload support on the device
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The issue of EAP frames sent to group address (or the wrong address) has been
addressed in mac80211, so this hack is no longer needed
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This caches flows between MAC addresses on separate ports, including their VLAN
in order to bypass the normal bridge forwarding code.
In my test on MT7622, this reduces LAN->WLAN bridging CPU usage by 6-10%,
potentially even more on weaker platforms
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Had to update generic defconfig (make kernel_menuconfig CONFIG_TARGET=generic)
for this bump, but since that only modifies the target defined in .config,
and since that target also needed to be updated for unrelated reasons, manually
propagated the newly added symbol to the generic config.
Removed upstreamed:
pending-5.10/860-Revert-ASoC-mediatek-Check-for-error-clk-pointer.patch[1]
All other patches automatically rebased.
1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.10.99&id=080f371d984e8039c66db87f3c54804b0d172329
Build system: x86_64
Build-tested: bcm2711/RPi4B, mt7622/RT3200
Run-tested: bcm2711/RPi4B, mt7622/RT3200
Signed-off-by: John Audia <graysky@archlinux.us>
Support MT7530 PHY link change interrupts, and enable for MT7621.
For external MT7530, a GPIO IRQ line is required, which is
board-specific, so it should be added to each DTS. In case the
interrupt-controller property is missing, it will fall back to
polling mode.
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
Add support for MediaTek Gigabit Ethernet PHYs found in MT7530 and
MT7531. Fix some link up/down issues.
The errornous check for the PHY mode which broke things with MT7531
has been removed as suggested by patch
net: phy: mediatek: remove PHY mode check on MT7531
As a result, things are working fine now on MT7622+MT7531 as well.
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
Tested-by: Daniel Golle <daniel@makrotopia.org>
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
The kernel configuration allows us to select a default qdisc. Let's do this for
5.10 (as 5.4 is on its way out) and get rid of the hacky patch we've been
carrying.
Acked-by: Jo-Philipp Wich <jo@mein.io>
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
This reverts commit 8b4cba53a9.
This broke the mt7530 on Linksys e8450 (mt7622) for me.
[ 1.312943] mt7530 mdio-bus:00 lan1 (uninitialized): failed to connect to PHY: -EINVAL
[ 1.320890] mt7530 mdio-bus:00 lan1 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 0
[ 1.331163] mt7530 mdio-bus:00 lan2 (uninitialized): failed to connect to PHY: -EINVAL
[ 1.339085] mt7530 mdio-bus:00 lan2 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 1
[ 1.349321] mt7530 mdio-bus:00 lan3 (uninitialized): failed to connect to PHY: -EINVAL
[ 1.357241] mt7530 mdio-bus:00 lan3 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 2
[ 1.367452] mt7530 mdio-bus:00 lan4 (uninitialized): failed to connect to PHY: -EINVAL
[ 1.375367] mt7530 mdio-bus:00 lan4 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 3
[ 1.385750] mt7530 mdio-bus:00 wan (uninitialized): failed to connect to PHY: -EINVAL
[ 1.393575] mt7530 mdio-bus:00 wan (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 4
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This reverts commit 3f4301e123.
This broke the mt7530 on Linksys e8450 (mt7622) for me.
[ 1.300554] mt7530 mdio-bus:00: no interrupt support
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Support MT7530 PHY link change interrupts, and enable for MT7621.
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
Add support for MediaTek Gigabit Ethernet PHYs found in MT7530.
Fix some link up/down issues.
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
Use hardware to forward multicast traffic instead of trapping to the
host.
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
Allow MTU up to 2026 on mediatek, ramips/mt7621 targets.
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
