A buffer is split into multiple descriptors if it exceeds 16 KB.
Apply the same split for the skb head as well (to deal with corner cases
on fraglist support)
Signed-off-by: Felix Fietkau <nbd@nbd.name>
9d843334 Update bash_completion
23cb3f38 Update manual pages
1d682dcd Bump up version number to 1.33.0, LT revision to 31:0:17
601fbbb4 Update doc
f44aa246 Update AUTHORS
dd74a6dd Update manual pages
e959e733 src: Refactor utos
fb9a204d nghttpx: Fix compile error without mruby
cd096802 Update doc
7417fd71 nghttpx: Per-pattern not per-backend
2d1a981c Merge branch 'akonskarm-master'
45acc922 clang-format
214d0899 Merge branch 'master' of https://github.com/akonskarm/nghttp2 into akonskarm-master
31fd707d nghttpx: Fix broken healthmon frontend
9a2e38e0 fix code for reuse addr on asio client
d24527e7 Bump up LT revision due to v1.32.1 release
6195d747 nghttpx: Share mruby context if it is compiled from same file
fb97f596 nghttpx: Allocate mruby file because fopen requires NULL terminated string
0ccc7a77 nghttpx: Move blocked request data to request buffer for API request
32826466 nghttpx: Fix crash with API request
0422f8a8 nghttpx: Fix worker process crash with neverbleed write error
e329479a Merge pull request #1215 from nghttp2/mruby-per-backend
f80a7873 Merge branch 'akonskarm-reuse_addr'
866ac6ab add option reuse addr in local endpoint configuration of asio client
b574ae6a nghttpx: Support per-backend mruby script
de4fd7cd doc: Update doc
32d7883c nghttpx: Downstream::request_buf_full: take into account blocked_request_buf_
9b24e197 nghttpx: Choose h1 protocol if headers have been sent to backend on retry
13ffece1 Merge pull request #1214 from nghttp2/fix-rst-without-dconn
9d5b781d Fix stream reset if data from client is arrived before dconn is attached
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
CAKE supports overriding of its internal classification of
packets through the tc filter mechanism.
Update the man page in our package, even though we don't
build them. Someone may find the documentation useful.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 30598a05385b0ac2380dd4f30037a9f9d0318cf2)
Expand filter flow mapping to include hosts as well
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit d14ffdc307d36bd9abe908b46ff7baece54c9551)
OpenWrt used to ship hardcoded defaults for lcp-echo-failure and
lcp-echo-interval in the non-uci /etc/ppp/options file.
These values break uci support for *disabling* LCP echos through
the use of "option keepalive 0" as either omitting the keepalive
option or setting it to 0 will result in no lcp-echo-* flags
getting passed to the pppd cmdline, causing the pppd process to
revert to the defaults in /etc/ppp/options.
Address this issue by letting the uci "keepalive" option default
to the former hardcoded values "5, 1" and by removing the fixed
lcp-echo-failure and lcp-echo-interval settings from the
/etc/ppp/options files.
Ref: https://github.com/openwrt/luci/issues/2112
Ref: https://dev.archive.openwrt.org/ticket/2373.html
Ref: https://bugs.openwrt.org/index.php?do=details&task_id=854
Ref: https://bugs.openwrt.org/index.php?do=details&task_id=1259
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This reverts commit 3a8efaef00.
The change reportedly breaks UART2 on some boards. Furthermore it uses
bitwise logic on an uninitialized variable and fails to explain what it
is fixing exactly.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The changed dictionary size leads to a different LZMA header which breaks
sysupgrade image magic checkibng on at least some RT288x boards.
Since the commit message only mentions testing on MT7621 and since the
change appears to break at least one other ramips subtarget, do not take
any chances and restrict the size limitation to only MT7621.
Fixes FS#1797
Fixes 09b6755946 ("ramips: limit dictionary size for lzma compression")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The rt3883 doesn't have a pinmux group named spi_cs1. The cs1 is part
of the pci group. The function pci-func enables the second chip select.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The PCI pins need to be set to "PCI Host support one device" to allow
the use of one PCI device and flash memory.
The pci-fnc function is intended to be used if no PCI is used but
flash, nand or the codec functionality is.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Previous refactoring of the script moved the LDSO detection into a
file-not-exists condition, causing onyl the very first executable to
get bundled.
Solve the problem by unconditionally checking for LDSO again.
Fixes: 9030a78a71 ("scripts: bundle-libraries: prevent loading host locales")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The control device /dev/cdc-wdm0 is not available immediately on the
D-Link DWR-921 Rev.C3, therefore the wwan interface fails to start at
boot with a "The specified control device does not exist" error.
This patch alters /lib/netifd/proto/qmi.sh to wait for
network.wwan.delay earlier, before checking for the control device,
instead of just before interacting with the modem.
One still has to use network.wwan.proto='qmi', as the "wwan" proto
performs that sort of check before any delay is possible, failing with a
"No valid device was found" error.
Signed-off-by: Thomas Equeter <tequeter@users.noreply.github.com>
This remedies an issue with the MBL Duo if both disks are inserted
and contain OpenWrt. kernel and dtb would be loaded from SATA 1:1
while rootfs (/dev/sda2) would be mounted on SATA 0:1.
Such a mix&match would obviously only work if both OpenWrt versions/
builds are identical, and especially fail after sysupgrade upgraded
the system disk on SATA 0:1.
The fallback to SATA 1:1 needs to be kept for MBL Single (only has
SATA 1:1) and MBL Duo with one disk inserted on SATA 1:1. To speed
up booting in those cases, the unneccesarily doubled "sata init"
will only be called once. (In theory it could be omitted completely
since the on-flash boot script already initializes SATA to load the
on-disk boot script.)
Tested on MBL Duo (all possible combination of disks) and MBL Single
Acked-by: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: Freddy Leitner <hello@square.wf>
The LEDs should be triggered/lit by any kind of state change instead of
turned on/off unconditional.
If LEDs really need to be turned off by default, it should be done via
the default-state devicetree led property.
The handling of the wndr3700v5 and wt3020 power led is at least
strange. Something is for sure wrong with them. Either the leds are
misnamed, the default off trigger is a typo or the polarity of the
gpios is wrong. Drop the power led from userspace and wait for someone
with access to the hardware to fix it properly.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Based on the userspace led configuration it's quite obvious that the
4g-0 led should be used for boot status indication.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Use the default-state property to express the desired led handling in
the devicetree source file instead of the userspace.
Signed-off-by: Mathias Kresin <dev@kresin.me>
All the LEDs are turned on by diag.sh at the end of the boot process.
No need to do the same via userspace configuration again.
Signed-off-by: Mathias Kresin <dev@kresin.me>
All boards either have a multi colour led or a single lightpipe. It
makes it impossible to handle the LEDs individual. Change the LED
config for these boards to take it into account.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Some combination of modem/wireless operator requires more time to
execute the commands.
Tested on DWR-512 embedded wwan modem and italian operator iliad (new
virtual operator).
Signed-off-by: Giuseppe Lippolis <giu.lippolis@gmail.com>
Set the window title not only in "xterm", but also in
e.g. "xterm-256color", "xterm-color", etc.
The case statement is taken from Debian / Ubuntu.
Signed-off-by: Paul Wassi <p.wassi@gmx.at>
This is necessary to get my position right.
Without this my longitude is incorrecty -15.85xxxx instead of -16.52yyyy
Signed-off-by: Bruno Randolf <br1@einfach.org>
Commit 7af1fb9faa ("kernel: add a RPS balancer") introduces a RPS balancer
for all targets.
In the past however, this patch was already introduced for target "mediatek"
in commit 7762c07c88 ("mediatek: bump to v4.14")
Remove the separate copy of the patch within the mediatek target,
which otherwise is applied twice and results in a build error.
Fixes: 7af1fb9faa ("kernel: add a RPS balancer")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Tested on 8devices Jalapeno(ipq40xx)
Introduces following changes:
Feature: Add support for WAKE_FILTER (WoL using filters)
Feature: Add support for action value -2 (wake-up filter)
Fix: document WoL filters option also in help message
Feature: ixgbe dump strings for security registers
Signed-off-by: Robert Marko <robimarko@gmail.com>
By default the RPS delegation will happen by masking the last few bits of
skb->hash. This patch adds an inermediate hash bucket that maps the masked
hash to a RPS core. This makes RPS results much more deterministic on SMP
systems.
Signed-off-by: John Crispin <john@phrozen.org>
The current make-ras.sh image generation script for the ZyXEL NBG6617
has portability issues with bash. Because of this, factory images are
currently not built correctly by the OpenWRT buildbots.
This commit replaces the make-ras.sh by C-written mkrasimage.
The new mkrasimage is also compatible with other ZyXEL devices using
the ras image-format.
This is not tested with the NBG6616 but it correctly builds the
header for ZyXEL factory image.
Signed-off-by: David Bauer <mail@david-bauer.net>
Currently all PCI devices get the same IRQ that affects performance badly.
This commit adresses this problem and cleans the code.
ar7100 has a special PCI interrupt controller@18060018 that works exactly
the same way as misc interrupt controller.
This patch does the following:
1. Defines pci-intc interrupt controller@18060018 in dtsi.
2. Removes interrupt-controller property from PCI node.
3. Sets a correct interrupt mask for PCI devices.
4. Removes all IRQ handling code from the PCI driver.
"qca,ar7100-misc-intc" should be used as the compatible property, becuase on ar7100
the controlled status register is read-only and the ack method used in
"qca,ar7240-misc-intc" won't work properly.
There are two very minor downsides of this patch that don't affect perormance:
1. We allocate an IRQ domain of 32 IRQ, whan we need only 5. But ar7100 aren't tiny un terms of RAM
and that is not very important and can be tuned if we implement "nr-interrupts" property".
2. It reuses the same irg chip name "MISC" for both controllers.
Run tested on DIR-825 B1.
Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
We currently don't have any code configuring interface mode in ath79,
meaning that we relies on bootloader to set the correct interface mode.
This patch added code to set interface correctly so that everything works
even if bootloader configures it wrong.(e.g. on WNDR3800 u-boot set
the second GMAC mode to RMII but it should be RGMII.)
Introduced "qca,mac-idx" for the difference in MII_CTRL register value.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Netgear WNR612v2 flashed with recent OpenWrt builds suffers from kernel
panic at boot during wireless chip initialization, making device
unusable:
ath: phy0: Ignoring endianness difference in EEPROM magic bytes.
ath: phy0: Enable LNA combining
CPU 0 Unable to handle kernel paging request at virtual address 1000fee1, epc == 801d08f0, ra == 801d0d90
Oops[#1]:
CPU: 0 PID: 469 Comm: kmodloader Not tainted 4.9.120 #0
[ ... register dump etc ... ]
Kernel panic - not syncing: Fatal exception
Rebooting in 1 seconds..
This simple patch fixes above error. It keeps LED table in memory after
kernel init phase for ath9k driver to operate correctly (__initdata
removed).
Also, another bug is fixed - correct array size is provided to function
that adds platform LEDs (this device has only 1 connected to Wifi chip)
preventing code from going outside array bounds.
Fixes: 1f5ea4eae4 ("ar71xx: add correct named default wireless led by using platform leds")
Signed-off-by: Michal Cieslakiewicz <michal.cieslakiewicz@wp.pl>
[trimmed commit message]
Signed-off-by: Mathias Kresin <dev@kresin.me>
The NBG6616 shares a config symbol with the NBG6716. It was accidentally
removed from the config when the ar71xx-tiny target was split off.
Fixes: 0cd5e85e7a ("ar71xx: create new ar71xx/tiny subtarget for 4MB flash devices")
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
This adds a shared ar7161_ubnt_routerstation.dtsi as well as two other
.dts files that utilize it, ar7161_ubnt_routerstation.dts and
ar7161_ubnt_routerstation-pro.dts.
The modifications to generic-ubnt.mk, config-default, and base-files
necessary for image generation, parsing RedBoot FIS partitions, network
configuration, and sysupgrade are also included.
This reintroduces vital bits from platform_do_upgrade_combined() and its
supporting functions to /lib/upgrade/platform.sh, which were previously
removed from ath79 in 3e9d9f6225 "ath79:
sysupgrade: drop unused platform checks". The new function is called
"routerstation_do_upgrade" and will *only* work for the RouterStation
series of boards. It does however retain the ability to downgrade (e.g.
from master -> 17.01.x using sysupgrade -F).
All hardware is functional including the AR8216 switch (for the Pro),
wireless via ath5k/ath9k using the miniPCI slots, flash, USB, button,
and LED.
Switch and LAN/WAN configuration is the same as it is with the
equivalent ar71xx targets. MAC addresses are assigned based upon the
content stored in the RedBoot config partition.
Flashing via both sysupgrade and TFTP has been confirmed to work. Also,
the initramfs images are now raw .bin files instead of being wrapped in
a uImage (as they currently are in ar71xx), which makes them bootable
with RedBoot.
One notable difference to ar71xx is the inclusion of the RedBoot
"fconfig" utility (analogous to U-Boot’s fw_printenv/fw_setenv) in
DEVICE_PACKAGES. The FIS partitions are probed using the RedBoot MTD
parser’s DT binding, whose proper usage is mutually exclusive to
defining a separate fixed-partitions node for "RedBoot config". This
config partition contains the board's base MAC address. The lack of a
hard-coded flash location means that the mtd-mac-address property cannot
be used in the .dts, so instead fconfig is used to read the MAC
addresses from flash in userspace during first boot.
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
Signed-off-by: Matt Merhar <mattmerhar@protonmail.com>
- fix single spaces hidden by a tab
- replace indentation with spaces by tabs
- make empty lines empty
- drop trailing whitespace
- drop unnecessary blank lines
Signed-off-by: Mathias Kresin <dev@kresin.me>
Signed-off-by: Paul Wassi <p.wassi@gmx.at>
The mt7620 doesn't have a pinmux group named spi_cs1. The cs1 is part
of the "spi refclk" group. The function "spi refclk" enables the second
chip select.
On reset, the pins of the "spi refclk" group are used as reference
clock and GPIO.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Binary patch the bundled glibc library to inhibit loading of host locale
archives in order to avoid triggering internal libc assertions when
invoking shipped, bundled executables.
The problem has been solved with upstream Glibc commit
0062ace229 ("Gracefully handle incompatible locale data") but we still
need to deal with older Glibc binaries for some time to come.
Fixes FS#1803
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
4c76aaee Update manual pages
2b51ad67 Bump up version number to 1.32.1, LT revision to 30:3:16
708379dc Tweak nghttp2_session_set_stream_user_data
73106b0d Compile with clang-6.0
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
