There is a problem with the EA8500, the switch will not work after soft
reboot, the only way to get it working again is to power cycle it
manually.
There are probably several issues in the play, it's quite hard to fix it
without having access to the actual device, so I don't see any other
option now, then revert the offending commit.
Ref: PR#2047
Fixes: FS#2168 ("Switch no longer work after restart on Linksys EA8500")
Reported-by: Adam <424778940z@gmail.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 04d6753d03)
The release notes since last time for wave-1:
* October 5, 2019: Fix too-short msg caused by invalid use of PayloadLen in receive path.
This appears to resolve the issue of getting (and ignoring) too-short commands
when we detect loss of CE interrupts and go into polling mode.
* October 12, 2019: Fix regression in IBSS mode that caused SWBA overrun issues. Related to
regression added during the ct-station logic, specifically TSF allocation.
Thanks for Ahmed Zaki @ Mage-Networks for helping to diagnose and test.
* October 15, 2019: Only send beacon tx completion events if we can detect CT driver is being
used (based on CT_STATS_OK flag being set). This should help CT firmware work
better on stock driver.
The release notes since last time for wave-2:
* October 15, 2019: Only send beacon tx completion events if we can detect CT driver is being
used (based on ATH10k_USE_TXCOMPL_TXRATE2 | ATH10k_USE_TXCOMPL_TXRATE1 flags being set).
This should help CT firmware work better on stock driver.
* October 31, 2019: Compile out peer-ratecode-list-event. ath10k driver ignores the event.
* November 1, 2019: Fix rate-ctrl related crash when nss and other things were changed while
station stays associated. See bug: https://github.com/greearb/ath10k-ct/issues/96
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit e716e93a2f7290086f49992c9980773c88100c3a)
Import patches from upstream to sync 19.07 with master:
9f3e3323e996 rt2x00: allow to specify watchdog interval
2034afe4db4a rt2800: add helpers for reading dma done index
759c5b599cf4 rt2800: initial watchdog implementation
09db3b000619 rt2800: add pre_reset_hw callback
710e6cc1595e rt2800: do not nullify initialization vector data
e403fa31ed71 rt2x00: add restart hw
0f47aeeada2a rt2800: do not enable watchdog by default
41a531ffa4c5 rt2x00usb: fix rx queue hang
3b902fa811cf rt2x00usb: remove unnecessary rx flag checks
1dc244064c47 rt2x00: no need to check return value of debugfs_create functions
706f0182b1ad rt2800usb: Add new rt2800usb device PLANEX GW-USMicroN
95844124385e rt2x00: clear IV's on start to fix AP mode regression
567a9b766b47 rt2x00: do not set IEEE80211_TX_STAT_AMPDU_NO_BACK on tx status
14d5e14c8a6c rt2x00: clear up IV's on key removal
13fa451568ab Revert "rt2800: enable TX_PIN_CFG_LNA_PE_ bits per band"
--pending-- rt2800: remove errornous duplicate condition
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
With this commit, the WAN LED is triggered by the switch port state
instead of the eth0 netdev.
Otherwise, the LED is always illuminated, regardless of the WAN port
link state.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 70d5989c9c)
Signed-off-by: David Bauer <mail@david-bauer.net>
This enables PMKSA and opportunistic key caching by default for
WPA2/WPA3-Personal, WPA3-Personal and OWE auth types.
Otherwise, Apple devices won't connect to the WPA3 network.
This should not degrade security, as there's no external authentication
provider.
Tested with OCEDO Koala and iPhone 7 (iOS 13.1).
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 3034f8c3b8)
Signed-off-by: David Bauer <mail@david-bauer.net>
Both targets miss a subtarget causing an image naming style which is
different from other all othe targets, even tho it already uses
`x/generic/` as subfolder as if the subtarget would exist.
This commit adds the Generic subtarget resulting in consistent naming.
~/src/openwrt/openwrt/bin/targets/ipq806x/generic$ ls
openwrt-ipq806x-generic-netgear_d7800-initramfs-uImage
openwrt-ipq806x-generic-netgear-d7800.manifest
openwrt-ipq806x-generic-netgear_d7800-squashfs-factory.img
openwrt-ipq806x-generic-netgear_d7800-squashfs-sysupgrade.bin
CC: John Crispin <john@phrozen.org>
Signed-off-by: Paul Spooren <mail@aparcar.org>
(cherry picked from commit 853e4dd306)
The 2.4 GHz radio had very poor signal reception (-89 dBm for an AP
sitting 5 m away). By enabling the external amplifier, received signal
has improved to -50 dBm for the same AP.
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
(cherry picked from commit e667d6f46b)
Modify GL-AR300M-Lite and GL-AR300M (NOR):
* Include qca9531_glinet_gl-ar300m.dtsi directly
rather than qca9531_glinet_gl-ar300m-nor.dts
* Remove redundant inclusion of gpio.h and input.h
Signed-off-by: Jeff Kletsky <git-commits@allycomm.com>
Reviewed-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit f5c7fe2ff0)
So far, WiFi MAC addresses for this device have been set up from
caldata. However, this returns values which do not look like MAC
addresses. They also do not match stock firmware:
wlan0 (5.0): 00:11:22:00:17:D0 from 0x8004
wlan1 (2.4): 00:11:22:00:17:CD from 0x4 (and 0x2e)
It looks like the only valid MAC address on this device is at 0x28.
So, this patch changes setup to calculate addresses based on the
value at 0x28:
lan: *:0A (flash, label)
wan: *:0B (flash + 1)
wifi2: *:0A (flash)
wifi5: *:0C (flash + 2)
Thanks to Roger Pueyo Centelles <roger.pueyo@guifi.net> for
investigating this on his devices.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit d1072096f4)
The kconfig symbol is an invisible one since its introduction. It is
not supposed to be enabled on its own.
Resolves FS#1821
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
(cherry picked from commit 4bf9bec361)
Improves rate control responsiveness and performance
Signed-off-by: Felix Fietkau <nbd@nbd.name>
[reworked to apply on 4.19.79 mac80211 + renumbered + refreshed]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
The Unifi AC-LR has identical hardware to the Unifi AC-Lite.
The antenna setup is different according to the vendor,
which explains the thicker enclosure.
Therefore, it is helpful to know the exact device variant,
instead of having "Ubiquiti UniFi-AC-LITE/LR".
Signed-off-by: Andreas Ziegler <dev@andreas-ziegler.de>
[fix legacy name in commit message; add old boardname to
SUPPORTED_DEVICES]
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 07c1ddf522)
Signed-off-by: David Bauer <mail@david-bauer.net>
This commit adds correct model detection for UniFi
AC-LR. Previously, said device was incorrectly detected
as UniFi-AC-LITE/MESH.
The Information about the device is stored at 0xC in the EEPROM
partition. It corresponds to the sysid in /etc/board.info of the
Ubiquiti stock firmware.
Signed-off-by: Andreas Ziegler <dev@andreas-ziegler.de>
[adjust naming style of target to existing ones]
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 2bc7c519dc)
Signed-off-by: David Bauer <mail@david-bauer.net>
This fixes frequent crashes observed on a UniFi AC Mesh using OpenWrt
master and 19.07. 18.06 seems not affected from our testing.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 641a93f0f2)
Signed-off-by: David Bauer <mail@david-bauer.net>
This commit changes the source of the Wave 1 ath10k-firmware
from linux-firmware to Kall Valos ath10k-firmware repository.
This is necessary as the firmware selected in linux-firmware produces
frequent crashes in some circumstances.
This patch can be removed as soon as linux-firmware carries
10.2.4-1.0-00047 firmware.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit a3914783a3)
Signed-off-by: David Bauer <mail@david-bauer.net>
Several Archer Cxx devices were using board-specific LED names in
ar71xx, which were changed to "tp-link:*" in ath79.
This patch adds migration for them.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 53e6cc7a81)
Several devices added to LED migration script will just have their
(old) board name converted to tp-link.
By using a variable for this, the amount of code in the migration
script can be reduced and the chance for typos is reduced.
This patch also introduces the marker for beginning of a pattern
"^" to the regex, so the match is more specific.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 6b0eb84336)
The code line patching ath9k MAC address for this device contains
a wrong number of arguments including an unset "$mac", which
looks like a typo or copy/paste mistake.
This has been introduced already in the device support commit
745dee11ac ("ath79: add support for WD My Net Wi-Fi Range
Extender").
This patch just removes the "$mac" argument, leaving a formally
valid line. (No on-device test has been performed.)
Cc: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 6b53033783)
The ar71xx images for the Ubiquiti NanoStation M (XM) devices use
"nanostation-m" as board name, but the ath79 images are only
compatible with the "nano-m" board name, so sysupgrade complains.
By changing this additional supported device, sysupgrade smoothly
upgrades from ar71xx to ath79.
Ref: openwrt#2418
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit f473ce6f23)
The ar71xx images for the Ubiquiti NanoStation M (XW) devices use
"nanostation-m-xw" as the board name, but the ath79 images are only
compatible with the "nano-m-xw" board name, so sysupgrade complains.
By adding this additional supported device, sysuspgrade smoothly
upgrades from ar71xx to ath79.
Tested on a NanoStation M (XW) running OpenWrt ar71xx r10250-016d1eb.
Ref: https://github.com/openwrt/openwrt/pull/2418
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
[removed duplicate DEVICE_VARIANT, removed uneeded nano-m-xw support]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 6dda2ea6ad)
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
The device did not appear to be reachable unless the connection were
forced to 100Mb or lower. Revert to previously working pll-data.
Also fix the phy-mode to represent the actual state needed for ethernet
to function.
Reported-by: Moritz Schreiber <moritz@mosos.de>
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
[add remark about phy-mode property]
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit ee41b602a2)
edma_read_append_stats() gets called from two places in the driver.
The first place is the kernel timer that periodically updates
the statistics, so nothing gets lost due to overflows.
The second one it's part of the userspace ethtool ioctl handler
to provide up-to-date values.
For this configuration, the use of spin_lock() is not sufficient
and as per:
<https://mirrors.edge.kernel.org/pub/linux/kernel/people/rusty/kernel-locking/c214.html>
the locking has to be upgraded to spin_lock_bh().
Signed-off-by: Masafumi UTSUGI <mutsugi@allied-telesis.co.jp>
[folded patch into 710-, rewrote message]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit f1d761f95e)
This adds engine configuration sections to openssl.cnf, with a commented
list of engines. To enable an engine, all you have to do is uncomment
the engine line.
It also adds some useful comments to the devcrypto engine configuration
section. Other engines currently don't have configuration commands.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit cebf024c4d)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
This corrects the additional boardname for the image metadata to the one
used in ar71xx. The previously present additional entry was never used
on a running system.
Reviewed-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 1a256470e7)
Signed-off-by: David Bauer <mail@david-bauer.net>
This commit fixes TP-Link TL-WDR4900 v1 MAC address assignment.
Previously, the MAC addrss was read for the ethernet from the "config"
partition. However, the content of this partition is dependent on the
firmware which was previously installed on the device.
Switch the MAC address source to the U-Boot partition, where the MAC
address is always present at a fixed partition. The partition was
previously already used for the WiFi MAC-addresses.
Reviewed-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 07e555d873)
Signed-off-by: David Bauer <mail@david-bauer.net>
The default sizes render Device/linksys_audi mage un-flashable.
Restore the pagesize, subpagesize, and blocksize for linksys_audi
from https://github.com/openwrt/archive.
Signed-off-by: Eubert Bao <bunnier@gmail.com>
(cherry picked from commit e11fc8439c)
The first allows usage of several functions in the std namespace, which
broke compilation of gddrescue specifically with uClibc-ng and uClibc++.
The second allows usage of long long with normal C++11, which is part of
the standard. Before, std=gnu++11 needed to be passsed to work around it.
As a result of the second patch, the pedantic patch can safely be removed.
Both patches are upstream backports.
Added -std=c++11 to CFLAGS to guarentee proper inclusion of long long.
Added another patch that fixes a typo with the long long support. Sent to
upstream.
Fixed up license information according to SPDX.
Small cleanups for consistency.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 6ab386c9bc)
Upon writing to "remove" file, debugfs_remove_recursive() blocks while
holding rtnl_lock. This is because debugfs' file_ops callbacks are
executed in debugfs_use_file_*() context which prevents file removal.
Fix this by only flagging the device for removal and then do the cleanup
in file_ops.release callback which is executed out of that context.
Signed-off-by: Ali MJ Al-Nasrawy <alimjalnasrawy@gmail.com>
(cherry picked from commit c2635b871d)
Since v3.11, netdevice notification data are of type
"struct netdev_notifier_info". Handle it as such!
This should fix a critical bug in which devices are unable get released
because trelay does not release resources in response to UNREGISTER
event spamming the log with something like:
unregister_netdevice: waiting for eth0.1 to become free. Usage count = 1
Signed-off-by: Ali MJ Al-Nasrawy <alimjalnasrawy@gmail.com>
(cherry picked from commit 77cfc0739d)
if gcc not linker whith this LDFLAGS, "file libbz2.so.1.0.8" will
recognize as pie executable ELF file ( which should be shared object).
this because the file command version before 5.36 not recognize
correctly.
Signed-off-by: leo chung <gewalalb@gmail.com>
(cherry picked from commit 56ab58fb6c)
This fixes a compilation error as follows:
drivers/staging/fsl_qbman/qman_config.c:815:29: error: bitwise comparison
always evaluates to false [-Werror=tautological-compare]
if ((qman_ip_rev & 0xFF00) == QMAN_REV31) {
Signed-off-by: Biwen Li <biwen.li@nxp.com>
(cherry picked from commit 53b73131e9)
The prepare target was added some 11 years ago to build tools and
toolchain and was recently extended to create buildinfo files for
reproducibility, meaning {feeds,version,config}.buildinfo.
As the buildbot workflow is more complex than the single prepare (kmod
feed insertion), prepare is only used to create those buildinfo files.
Running prepare however runs `target/compile` as well, taking time even
everything is already compiled.
Splitting this allows the buildbot to run only the `buildinfo` target
while others can still use the convenience feature `prepare`.
Signed-off-by: Paul Spooren <mail@aparcar.org>
(cherry picked from commit 6caf437652)
The destination buffer size `d_len` is passed to `lzma_inflate` as a
pointer. Therefore, it needs to be dereferenced to compare its content.
Signed-off-by: Christian Franke <nobody@nowhere.ws>
(cherry picked from commit d544bc84a0)
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
The fwutil command will interpret the final 16 byte of a given firmware
image files as "struct fwimage_trailer".
In case these bytes do look like a valid trailer, we must ensure that we
print them out along with the remainder of the image to not accidentally
truncate non-trailer-images by 16 bytes when they're piped through fwtool,
e.g. as part of an image verification command sequence.
Some command sequences pipe images through fwtool in order to strip any
possible metadata, certificate or signature trailers and do not expect
bare images without any of that metadata to get truncated as other non-
fwtool specific metadata is expected at the end of the file, e.g. an
information block with an md5sum in case of the combined image format.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 889b841048)
The switch port naming in LuCI does not fit the physical numbers
on the front of this device. Since this is confusing, fix it.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit e56e5a454e)
GDB 8.3.1 brings the following fixes and enhancements over GDB 8.3:
PR c++/20020 (GDB segfault on printing objects)
PR gdb/24454 (nat/x86-linux-dregs.c failed assertion)
PR breakpoints/24541 (Incorrect evaluation of systemtap probes due to register being signed and probe expression assuming unsigned)
PR symtab/24545 (Symbol loading performance regression with cc1)
PR gdb/24592 (amd64->i386 linux syscall restart problem)
PR gdb/25009 (terminate called after throwing an instance of 'srchilite::ParserException')
PR gdb/25010 (Calls to error () can cause SIGTTOU to send gdb to the background)
PR breakpoints/25011 (Breakpoints on file reloads broken for PIE binaries)
This corrective release also brings the following testsuite fixes and
enhancements:
PR testsuite/25005 (gdb-caching-proc.exp takes a lot of time on skip_opencl_tests)
PR testsuite/25016 (Test-case failures for -pie)
GDB 8.3 includes the following changes and enhancements:
* Support for new native configurations (also available as a target configuration):
- RISC-V GNU/Linux (riscv*-*-linux*)
- RISC-V FreeBSD (riscv*-*-freebsd*)
* Support for new target configurations:
- CSKY ELF (csky*-*-elf)
- CSKY GNU/Linux (csky*-*-linux)
- NXP S12Z ELF (s12z-*-elf)
- OpenRISC GNU/Linux (or1k*-*-linux*)
* Native Windows debugging is only supported on Windows XP or later.
* The Python API in GDB now requires Python 2.6 or later.
* GDB now supports terminal styling for the CLI and TUI.
Source highlighting is also supported by building GDB with GNU
Highlight.
* Experimental support for compilation and injection of C++ source
code into the inferior (requires GCC 7.1 or higher, built with
libcp1.so).
* GDB and GDBserver now support IPv6 connections.
* Target description support on RISC-V targets.
* Various enhancements to several commands:
- "frame", "select-frame" and "info frame" commands
- "info functions", "info types", "info variables"
- "info thread"
- "info proc"
- System call alias catchpoint support on FreeBSD
- "target remote" support for Unix Domain sockets.
* Support for displaying all files opened by a process
* DWARF index cache: GDB can now automatically save indices of DWARF
symbols on disk to speed up further loading of the same binaries.
* Various GDB/MI enhancements.
* GDBserver on PowerPC GNU/Linux now supports access to the PPR,
DSCR, TAR, EBB/PMU, and HTM registers.
* Ada task switching support when debugging programs built with
the Ravenscar profile added to aarch64-elf.
* GDB in batch mode now exits with status 1 if the last executed
command failed.
* Support for building GDB with GCC's Undefined Behavior Sanitizer.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
GDB 8.3.1 brings the following fixes and enhancements over GDB 8.3:
PR c++/20020 (GDB segfault on printing objects)
PR gdb/24454 (nat/x86-linux-dregs.c failed assertion)
PR breakpoints/24541 (Incorrect evaluation of systemtap probes due to register being signed and probe expression assuming unsigned)
PR symtab/24545 (Symbol loading performance regression with cc1)
PR gdb/24592 (amd64->i386 linux syscall restart problem)
PR gdb/25009 (terminate called after throwing an instance of 'srchilite::ParserException')
PR gdb/25010 (Calls to error () can cause SIGTTOU to send gdb to the background)
PR breakpoints/25011 (Breakpoints on file reloads broken for PIE binaries)
This corrective release also brings the following testsuite fixes and
enhancements:
PR testsuite/25005 (gdb-caching-proc.exp takes a lot of time on skip_opencl_tests)
PR testsuite/25016 (Test-case failures for -pie)
GDB 8.3 includes the following changes and enhancements:
* Support for new native configurations (also available as a target configuration):
- RISC-V GNU/Linux (riscv*-*-linux*)
- RISC-V FreeBSD (riscv*-*-freebsd*)
* Support for new target configurations:
- CSKY ELF (csky*-*-elf)
- CSKY GNU/Linux (csky*-*-linux)
- NXP S12Z ELF (s12z-*-elf)
- OpenRISC GNU/Linux (or1k*-*-linux*)
* Native Windows debugging is only supported on Windows XP or later.
* The Python API in GDB now requires Python 2.6 or later.
* GDB now supports terminal styling for the CLI and TUI.
Source highlighting is also supported by building GDB with GNU
Highlight.
* Experimental support for compilation and injection of C++ source
code into the inferior (requires GCC 7.1 or higher, built with
libcp1.so).
* GDB and GDBserver now support IPv6 connections.
* Target description support on RISC-V targets.
* Various enhancements to several commands:
- "frame", "select-frame" and "info frame" commands
- "info functions", "info types", "info variables"
- "info thread"
- "info proc"
- System call alias catchpoint support on FreeBSD
- "target remote" support for Unix Domain sockets.
* Support for displaying all files opened by a process
* DWARF index cache: GDB can now automatically save indices of DWARF
symbols on disk to speed up further loading of the same binaries.
* Various GDB/MI enhancements.
* GDBserver on PowerPC GNU/Linux now supports access to the PPR,
DSCR, TAR, EBB/PMU, and HTM registers.
* Ada task switching support when debugging programs built with
the Ravenscar profile added to aarch64-elf.
* GDB in batch mode now exits with status 1 if the last executed
command failed.
* Support for building GDB with GCC's Undefined Behavior Sanitizer.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 942f020bf4d0f49e4c5586a59ddf21c00e9d8bbe)
1.1.24 release notes
new features:
- GLOB_TILDE extension to glob
- non-stub catgets localization API, using netbsd binary catalog format
- posix_spawn file actions for [f]chdir (extension, pending future standard)
- secure_getenv function (extension)
- copy_file_range syscall wrapper (Linux extension)
- header-level support for new linux features in 5.2
performance:
- new fast path for lrint (generic C version) on 32-bit archs
major internal changes:
- functions involving time are overhauled to be time64-ready in 32-bit archs
- x32 uses the new time64 code paths to replace nasty hacks in syscall glue
compatibility & conformance:
- support for powerpc[64] unaligned relocation types
- powerpc[64] and sh sys/user.h no longer clash with kernel asm/ptrace.h
- select no longer modifies timeout on failure (or at all)
- mips64 stat results are no longer limited to 32-bit time range
- optreset (BSD extension) now has a public declaration
- support for clang inconsistencies in wchar_t type vs some 32-bit archs
- mips r6 syscall asm no longer has invalid lo/hi register clobbers
- vestigial asm declarations of __tls_get_new are removed (broke some tooling)
- riscv64 mcontext_t mismatch glibc's member naming is corrected
bugs fixed:
- glob failed to match broken symlinks consistently
- invalid use of interposed calloc to allocate initial TLS
- various dlsym symbol resolution logic errors
- semctl with SEM_STAT_ANY didn't work
- pthread_create with explicit scheduling was subject to priority inversion
- pthread_create failure path had data race for thread count
- timer_create with SIGEV_THREAD notification had data race getting timer id
- wide printf family failed to support l modifier for float formats
arch-specific bugs fixed:
- x87 floating point stack imbalance in math asm (i386-only CVE-2019-14697)
- x32 clock_adjtime, getrusage, wait3, wait4 produced junk (struct mismatches)
- lseek broken on x32 and mipsn32 with large file offsets
- riscv64 atomics weren't compiler barriers
- riscv64 atomics had broken asm constraints (missing earlyclobber flag)
- arm clone() was broken when compiled as thumb if start function returned
- mipsr6 setjmp/longjmp did not preserve fpu register state correctly
Refreshed all patches.
Removed upstreamed.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Memory auto-detection for mt7621 has just been added to 19.07
stable branch.
This removes the memory node for the ZBT-WE1326, which will support
revision 5 that has 256MiB RAM (Nanya NT5CC128M16IP-DI) instead of
512MiB (up to revision 4).
ref: #1930
This is taken from master commit a2c19f1d2f ("ramips: dts: drop
memory nodes"), where _all_ memory nodes were removed.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
mt7621 has the following memory map:
0x0-0x1c000000: lower 448m memory
0x1c000000-0x2000000: peripheral registers
0x20000000-0x2400000: higher 64m memory
detect_memory_region in arch/mips/kernel/setup.c only add the first
memory region and isn't suitable for 512m memory detection because
it may accidentally read the memory area for peripheral registers.
This commit adds memory detection capability for mt7621:
1. add the highmem area when 512m is detected.
2. guard memcmp from accessing peripheral registers:
This only happens when some weird user decided to change
kernel load address to 256m or higher address. Since this
is a quite unusual case, we just skip 512m testing and return
256m as memory size.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
(cherry picked from commit 6d91ddf517)
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
`make xconfig` fails with following linking error of qconf binary:
g++ -lQt5Widgets -lQt5Gui -lQt5Core -o qconf qconf.o zconf.tab.o
/usr/bin/ld: qconf.o: in function ConfigList::metaObject() const': qconf.cc:(.text+0x3eb): undefined reference to QObjectData::dynamicMetaObject() const'
/usr/bin/ld: qconf.o: in function `ConfigList::qt_metacast(char const*)': link error.
which is caused by the wrong order of the linked objects/libraries so
this patch reorders the linker's arguments which makes the qconf compile
again.
Signed-off-by: leo chung <gewalalb@gmail.com>
[commit subject and message tweaks, whitespace fix]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit b2c55d50f8)
On a recent Gentoo Linux installation, invoking `make menuconfig`, `make
kernel_menuconfig` or `make kernel_nconfig` in the build system fails,
whereas for example `make menuconfig` in the kernel tree alone works as
expected.
This is happening because STAGING_PREFIX is not defined when kernel's
{menu,n}config target calls pkg-config from the toolchain/host and thus
pkg-config returns an empty value, and the fallback values in the kernel
config script are applied but those are off and the linking fails.
Solution is to use system's pkg-config for all ncurses based menu config
targets in order to provide proper compiler/linker flags.
Ref: FS#2423
Cc: Thomas Albers <thomas.gameiro@gmail.com>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
Tested-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 965f341aa9)
This fixes off-by-one error introduced in commit dc76900021
("kernel: Correctly search for the FIT image in mtd partition.")
Function `mtd_read` starts reading at `offset` and
needs `hdr_len` number of bytes to be available. Suppose
the easiest case when `offset` is `0` and `hdr_len` equals
to `mtd->size` - the `for` loop will not be entered even
when enough bytes are available to be read.
Same happens for any non-zero `offset`, when `hdr_len` is
just enough bytes to be read until `mtd->size` is reached.
Imagine that for example `mtd->size=5`, `offset=4` and
`hdr_len=1`. Then `offset+hdr_len=5` and the check has to
be `offset+hdr_len <= mtd->size`, i.e. `5 <= 5`. The
check for `offset + hdr_len` value needs to be inclusive,
therefore use `<=`.
Fixes: dc76900021 ("kernel: Correctly search for the FIT image in mtd partition.")
Signed-off-by: Oldřich Jedlička <oldium.pro@gmail.com>
[adjusted commit ref, fixes tag]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit e0ce80d42a)
Previously all iterations of the loop checked offset=0 in the partition.
Signed-off-by: Fredrik Olofsson <fredrik.olofsson@anyfinetworks.com>
(cherry picked from commit dc76900021)
If no device tree is given there is no node generated, but
the configuration does still include the name of the missing node.
This will result in a successful build fit image, but bootm does
throw a error message if we want to boot the bad configuration.
Signed-off-by: Johann Neuhauser <jneuhauser@dh-electronics.com>
(cherry picked from commit 205e0939f0)
The actual retrieval of the MAC address in mtd_get_mac_binary_ubi()
is the same as in get_mac_binary(). Thus, use the latter function
in the former to reduce duplicate code.
This will also allow to benefit from the enhanced path check there
and bring mtd_get_mac_binary_ubi() more in line with the similar
mtd_get_mac_binary().
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 45600124fc)
Now that the md5 check is fixed and metadata present, sysupgrade on
ar71xx will complain about device not being supported by the image.
Since the cause is not matching strings for supported devices add them
accordingly.
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
(cherry picked from commit a45cf75eca)
This target enforces metadata check so add the necessary information. It
was previously removed because md5 sum check. When using these sysupgrade
images on ar71xx target the check would complain about them not matching.
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
(cherry picked from commit 077d06a1a5)
There is md5 sum of whole image embedded in combined-image header which
is checked on sysupgrade. The check will fail for ath79 images which
may have embedded metadata. This is because metadata are appended after
the combined image is created. To allow smooth transition from ar71xx to
ath79, strip metadata before calculating md5 sum for whole image.
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
(cherry picked from commit 8b4109c2b4)
The frequency was filled acording the information from datasheet for
particular chip (Winbond 25Q128BVFG). Unfortunately this led to
coruption and introduced bad blocks on the chip. Reducing the frequency
to commonly used in ath79, made the board more stable and no new bad
blocks were spoted.
Fixes: b3a0c97 ("ath79: add support for jjPlus JA76PF2")
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
(cherry picked from commit c1db564cbc)
The relationship between GMAC0 and GMAC1 and the kernel devices
eth0 and eth1 was reversed for many ath79 devices by commit 8dde11d521
ath79: dts: drop "simple-mfd" for gmacs in SoC dtsi
The GL-AR300M-Lite is a single-port device, with the "LAN" port of the
GL-AR300M board unpopulated and its sole port now referenced as eth1,
as a result of commit 8dde11d521. The device was unreachable on
first boot or fresh config.
By changing ð1 (GMAC1) to an MFD, GMAC0 is able to associate with
the phy and is known by the kernel as "eth0".
Thanks to Chuanhong Guo for the suggestion of "simple-mfd"
Signed-off-by: Jeff Kletsky <git-commits@allycomm.com>
(cherry picked from commit b90ea19860853dd538e704e3e4402686c316e43c)
Previously, the board name for the GL-AR300M-Lite was incorrect
in 02_network, resulting in an unintended, fall-through condition
when initializing the network configuration.
While builds prior to commit 8dde11d521 (merged June 5, 2019)
ath79: dts: drop "simple-mfd" for gmacs in SoC dtsi
functioned properly, the error was noted in resolving first-boot
connectivity issues related to the single-phy nature of the device
and the "swap" of eth0 and eth1 related to that commit.
Signed-off-by: Jeff Kletsky <git-commits@allycomm.com>
(cherry picked from commit 549ea4dc1878d95ee2b752b8840d75a64a67b679)
The following symbols, selected by the qce driver were removed:
CONFIG_CRYPTO_CBC
CONFIG_CRYPTO_CTR
CONFIG_CRYPTO_DES
CONFIG_CRYPTO_DEV_QCE
CONFIG_CRYPTO_ECB
CONFIG_CRYPTO_NULL
CONFIG_CRYPTO_SEQIV
CONFIG_CRYPTO_XTS
CONFIG_CRYPTO_GF128MUL was removed as well, since it is only needed by
some cipher modes (LRW, GCM), none of which are selected, and it is
packaged as a module.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
This adds the CRYPTO_ALG_KERN_DRIVER_ONLY flag to Qualcomm crypto engine
driver algorithms, so that openssl devcrypto can recognize them as
hardware-accelerated.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
[refresh, move to ipq40xx as its the only target right now]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 391b14a892)
This patch removes a typo (extra "0") so that the 'cpu-alert6'
step is triggered once the system reaches 85°C.
Note: Unless the WNDR4700 is placed in an hot oven, the
hardware-monitor will never reach this value.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Replace an old cleanup patch that never made it upstream with the proper
upstream fix. This patch was incompatible with the recent changes that
affected the way that the flow tuple dst entry was used.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
(cherry-picked from commits 442ecce761 and c8933ce533)
Commit fcb41decf6 ("config: enable some useful features on
!SMALL_FLASH devices") enabled netns, which in turn lead to the crash in
the flow offload target.
When the flow offloading framework intends to delete a flow from the
hardware table, it is necessary to retrieve the namespace from
nf_flowtable->ft_net. However, no one ever wrote the namespace into
nf_flowtable->ft_net in advance. So the framework will mistakenly use a
NULL namespace to execute dev_get_by_index_rcu(net, ifindex), leading to
the kernel panic.
Ref: FS#2321
Fixes: fcb41decf6 ("config: enable some useful features on !SMALL_FLASH devices")
Tested-by: Simon Tretter <simon@mediaarchitectu.re>
Signed-off-by: HsiuWen Yen <y.hsiuwen@gmail.com>
[merged patch into offload patch, fix for 4.19, SOB fix, commit subj/msg touches]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry-picked from commit d344591e72)
$(board_name) was providing content on "boardtype" (and optionally
"boardnum") NVRAM values. That function requires & expects more specific
and detailed model name extracted from the /proc/cpuinfo.
Fixes: f12a32630f ("treewide: use the generic board_name function")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
572ff7f fritzcreator: actually add checksum spacer
6edce1a fritzcreator: replace obscure padding generation with something more portable
2ff189f add ASUS RT-AC58U "easy install" factory u-boot shim
b91f9c2 readd spi-nand support
486ae53 improve cmd_sysupgrade
b0933f1 replace sstrip with strip
882e48a do not include generated files into git
0c5aa5f fix bugs in ipq40xx_cdp.c
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit af63436d2d)
The ar40xx driver currently panics in case no QCA807x PHY has been
successfully probed. This happens when the external PHY is still
in reset when probing the ar40xx switch driver.
Note that this patch does not fix the root cause, ar40xx_probe now
simply fails instead of causing a kernel panic due to a nullpointer
dereference.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit e2c084cabc)
5e8cd86f90da ath10k-ct: Backport ap-vlan code from 5.2 to 4.20 and 4.19 drivers.
0c518586bd7f ath10k-ct: Fix a few warning splats.
Adds AP VLAN.
Refreshed all patches.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
This enables a feature flag in the wave-2 firmware wmi-services indicating it can send
software-encrypted raw frames. This should in turn allow the AP-VLAN feature to work.
Signed-off-by: Robert Marko <robimarko@gmail.com>
(cherry picked from commit 7c930990af)
We are not sure if 640-bridge-only-accept-EAP-locally.patch is still needed
as a first step, add disable_eap_hack sysfs config to allow to disable it
Signed-off-by: Etienne Champetier <champetier.etienne@gmail.com>
(cherry picked from commit 7d542dc804)
This version fixes 3 low-severity vulnerabilities:
- CVE-2019-1547: ECDSA remote timing attack
- CVE-2019-1549: Fork Protection
- CVE-2019-1563: Padding Oracle in PKCS7_dataDecode and
CMS_decrypt_set1_pkey
Patches were refreshed.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit d868d0a5d7)
Some ar9344-based devices are using ath9k_patch_fw_mac_crc, which
is meant to generate a checksum, for fixing their ath9k MAC
addresses.
However, those do not have a checksum field, and the calculated
checksum offset would be negative.
This patch will use ath9k_patch_fw_mac function for those devices.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit f8d8b3f85d)
This patch was originally added to fix compilation with v4l2rtspserver.
Turns out it was v4l2rtspserver that was broken, not uClibc++. This now
causes issues with a different package where the arguments are being
split.
Note that with this patch, shellcheck throws an error:
SC2068: Double quote array expansions to avoid re-splitting elements.
More: https://github.com/openwrt/packages/pull/9972#discussion_r324878373
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 977a8fc5fc)
Remove 300-bn_mul.h-Use-optimized-MULADDC-code-only-on-ARM-6.patch,
the issue has been fixed upstream.
Signed-off-by: Magnus Kroken <mkroken@gmail.com>
(cherry picked from commit 49d96ffc5c)
add module to support Emulex OneConnect
common in 10Gbit SFP+ cards by Dell/HP/IBM
supports OneConnect OCe10xxx OCe11xxx OCe14xxx,
LightPulse LPe12xxx
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
(cherry picked from commit 827f47749b)
Since service and instance names may contain characters which are not allowed
in JSON path labels, such as dashes or spaces, change the filter expression
to array square bracket notation to properly match these cases as well.
Fixes: 2c3dd70741 ("procd: add procd_running() helper for checking running state")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit c933b6d224)
This adds an easy-installation factory image for the NETGEAR R6220
router. The factory image can either be flashed via the vendor Web-UI or
the bootloader using nmrpflash.
Tested with NETGEAR V1.1.0.86 firmware.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 607dfdf211)
Allow flashing Luxul devices using vendor firmware format. The next step
will be building proper images once they are conirmed to work.
Signed-off-by: Dan Haab <dan.haab@legrand.com>
(cherry picked from commit bc5db7364d)
This makes brcmfmac use the same wiphy after PCIe reset to help user
space handle corner cases (e.g. firmware crash).
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit f39f4b2f6d)
Now that $UPGRADE_BACKUP is set conditionally there is no need to check
the $UPGRADE_OPT_SAVE_CONFIG anymore. All conditions can be simplified.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit a858db3136)
b8238df sysupgrade: support "backup" attribute
This update requires "sysupgrade" method callers to pass "backup"
attribute if $UPGRADE_BACKUP is used in the project.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 9785a9121d)
This explicitly tells procd what backup file should be used during
sysupgrade (if any). It's much more generic this way compared to the
magic /tmp/sysupgrade.tgz file that had to be created before a call.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit c5223b26a4)
The UniFi AC LED mapping is currently off. The blue/white LED are used
as WiFi indicators, while the vendor firmware does not feature WiFI
LEDs.
Instead, the LEDs are used to indicate the devices status. Align the LED
mapping to match the vendor firmware as good as possible.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 18fa749df8)
a88fb42 iwinfo: add device id for Qualcomm Atheros QCA9886
1b69d86 iwinfo: add device id for Qualcomm Atheros QCA9887
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 7db2f1a71f)
Move the IRQ fix from generic to ar71xx specific.
Other targets like ath79 have specific pathes to delete this code.
This resulted in a build failure on ath79
Fixes: 00d48bcac0 ("ar71xx: Fix potentially missed IRQ handling during
dispatch")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
If both interrupts are set in the current implementation
only the 1st will be handled and the 2nd will be skipped
due to the "if else" condition.
Fix this by using the same approach as done for QCA955x
just below it.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 9e8c36557cc0582986862f5a36e17adf6db2b90e)
If both interrupts are set in the current implementation
only the 1st will be handled and the 2nd will be skipped
due to the "if else" condition.
Fix this by using the same approach as done for QCA955x
just below it.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
This allows platform code to check if firmware image can be used with
preserving a backup. It may be used e.g. when installing vendor
firmwares that won't restore appended backup archive.
Suggested-by: Luis Araneda <luaraneda@gmail.com>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 1c510fe298)
This provides TRX validation result to the validation JSON. It also
prevents users from installing broken firmware files.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit c19b9f9a26)
This provides TRX validation result, so final JSON may look like:
{
"tests": {
"fwtool_signature": true,
"fwtool_device_match": true,
"trx_valid": true
},
"valid": true,
"forceable": true
}
It also prevents users from installing broken firmware files, e.g.:
root@OpenWrt:/# sysupgrade -F -n /tmp/TZ
Image metadata not found
Invalid image type. Please use firmware specific for this device.
Image check failed but --force given - will update anyway!
Commencing upgrade. Closing all shell sessions.
Firmware image is broken and cannot be installed
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit e68c1cebd1)
It's a variable set by procd that should replace hardcoded
/tmp/sysupgrade.tgz.
This change requires the most recent procd with the commit 0f3c136
("sysupgrade: set UPGRADE_BACKUP env variable").
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 641f6b6c26)
$CONF_TAR shouldn't be assumed to always point to the sysupgrade.tgz.
This change makes code more generic and allows refactoring $CONF_TAR.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 62dbe361a1)
1) Add BACKUP_FILE and use it when copying an archive to be restored
after sysupgrade (on the next preinit).
2) Use CONF_TAR for copying backup prepared by the /sbin/sysupgrade
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit bf39047872)
Including "sys/stat.h" from newer glibc will cause __u64 from linux uapi
header to be included, causing compilation failure for u-boot tools
USE_HOSTCC
Remove typedef for __u64 in include/compiler.h to fix the issue. It should be
safe because as of u-boot-2018.03, no ref to __u64 is found under u-boot tools/
directory
Error message snippet follows
HOSTCC tools/mkenvimage.o
In file included from /usr/include/asm-generic/types.h:7,
from /usr/include/asm/types.h:5,
from /usr/include/linux/types.h:5,
from /usr/include/linux/stat.h:5,
from /usr/include/bits/statx.h:30,
from /usr/include/sys/stat.h:446,
from tools/mkenvimage.c:21:
/usr/include/asm-generic/int-ll64.h:31:42: error: conflicting types for '__u64'
31 | __extension__ typedef unsigned long long __u64;
| ^~~~~
In file included from <command-line>:
././include/compiler.h:69:18: note: previous declaration of '__u64' was here
69 | typedef uint64_t __u64;
| ^~~~~
make[5]: *** [scripts/Makefile.host:116: tools/mkenvimage.o] Error 1
Ref: https://forum.openwrt.org/t/compile-error-19-07/44423
Ref: https://bugzilla.redhat.com/show_bug.cgi?id=1699194
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
This reverts commit af91a370de.
As Piotr Dymacz pointed out:
In QCA MIPS based WiSOCs, for first USB interface,
device/host mode can be selected _only_ in hardware
see description of 57c641ba6e
QCA955x and QCA9563, second USB can be switched to device
mode in software (tested and confirmed on real hardware).
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Current ar724x code does the reset only on single pci bus, and
in case of qca9558 writes the wrong register (0x10 vs 0x0c).
This change allows the reset of second pci bus, commonly used in
Archer C7 devices, in case host controller is stuck in reset.
If the resetting controller on boot can solve any other issue it
can be enabled unconditionally by removing reset check before
ar724x_pci_hw_init is called.
Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>
[refreshed to apply cleanly]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 76d870871cb12fc0c170e5fd23bce568adfaae6d)
Enable flushing of write buffers on qca955x. GPL code has 0x88 reg
defined for PCI flush which is likely an error since the device
freezes on boot. So use DS default value 0xA8 for PCI flush.
Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>
(cherry picked from commit fe9e702dc94ece2a004f6db68d6fb9a94d9437cb)
Switch from ci_usb_setup to generic platform initialization of
usb0 port.
Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>
(cherry picked from commit 36a0cfd24be1cb79f221964ed2bfe12b98befff3)
This backports the only non-cosmetic fix from 6640e1c368
("ramips: clean and improve MAC address setup in 02_network").
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
In 555ca422d1 ("ramips: fix D-Link DIR-615 H1 switch port
mapping"), port setup for dir-615-h1 was changed without removing
the old one. This was working as the new one was triggered earlier
than the old one.
(In the meantine, changed sorting during ramips rename patches
actually inversed that order.)
Anyway, just remove the wrong case now.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit e35e4a996e)
ARC FreeStation5 is present twice in MAC address setup.
>From older commits/changes, it is not possible to reconstruct
the correct choice only by reading the annotations.
Thus, remove the second case and keep the first one, so behavior
stays the same (as nobody seems to have complained about it).
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit ad4eb2241b)
Patch getting RAM info got upstreamed. A debugging fs entry for testing
reset feature was added.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 681acdcc54)
Image generation is currently failing on builbots due to the following
error:
WARNING: Image file openwrt-19.07-snapshot-r10495-db5164d3d0-ramips-mt7620-wt3020-4M-squashfs-factory.bin is too big
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This patch was applied to the upstream kernel in version 4.14.135,
remove it from our patches directory.
Fixes: 40379b0ec6 ("apm821xx: fix bogus key-presses on boot")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This reverts commit 1c0290c5cc.
Dropping the nopad can make the padding overflow into the next erase
block on devices using a non-aligned rootfs start. This breaks the jffs2
overlay partition with the following messages:
[ 30.343877] jffs2_scan_eraseblock(): End of filesystem marker found at 0x10000
[ 30.376512] jffs2: Cowardly refusing to erase blocks on filesystem with no valid JFFS2 nodes
[ 30.385253] jffs2: empty_blocks 196, bad_blocks 0, c->nr_blocks 197
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
(cherry picked from commit f11d90a76b)
The AVM Fritz!Box 7530 (and probably other AVM IPQ4019 NAND devices)
has it's caldata not stored consistently, but instead at currently
3 known possible offsets.
As we get a non-zero exit code from fritz_cal_extract, simply try all
three possible offsets on both bootloader partitions, until a matching
caldata for each radio is found.
Reported-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit a6f85b81b7)
The net pointer in struct xt_tgdtor_param is not explicitly
initialized therefore is still NULL when dereferencing it.
So we have to find a way to pass the correct net pointer to
ipt_destroy_target().
The best way I find is just saving the net pointer inside the per
netns struct tcf_idrinfo, which could make this patch smaller.
Fixes: 0c66dc1ea3f0 ("netfilter: conntrack: register hooks in netns when needed by ruleset")
Reported-and-tested-by: Tony Ambardar <itugrok@xxxxxxxxx>
Cc: Jamal Hadi Salim <jhs@xxxxxxxxxxxx>
Cc: Jiri Pirko <jiri@xxxxxxxxxxx>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
[Backport for kernel v4.19 and v4.14]
[Bug Link: https://bugzilla.kernel.org/show_bug.cgi?id=204681]
Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
(cherry picked from commit 7735cce0c5)
sysupgrade passes image path to platform_check_image() as an argument so
it can be simply accessed using $1
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 76e43c82b5)
[rmilecki: dropping ARGV without this change broke sysupgrade]
Fixes: 6ac62c4b6c ("base-files: don't set ARGV and ARGC")
stage2 passes image path to platform_do_upgrade() as an argument so it
can be simply accessed using $1
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 8b4bc7abe0)
[rmilecki: dropping ARGV without this change broke sysupgrade]
Fixes: 6ac62c4b6c ("base-files: don't set ARGV and ARGC")
This fixes the previously incorrect phy-mode for the OCEDO Ursus GMAC0.
See 62abbd587d ("ath79: correct various phy-mode properties")
for more details.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 7b137e9df9)
There are currently the following issues present for the Netgear R6220,
R6350 and WNDR3700 v5:
- LAN and WAN MAC-addresses are inverted
- WAN MAC-address is off. It are +2 compared to the LAN MAC-address
(R6350 only)
- Switchport order is inverted in LuCi
This commit fixes both these issues by assigning correct MAC-addresses
to LAN and WAN interfaces and defining the switchports with the correct
labels.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 13937a16d4)
This commit switches the default trigger for the WiFi LED from a netdev
trigger on "wlan0" to a wireless-phy based trigger. THis allows the LED
to work, even when the wireless interface is not named "wlan0" without
modifiying the LED settings.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit fa46c9b208)
The MediaTek MT7621 NAND driver currently intransparently shifts NAND
pages when a block is marked as bad. Because of this, offsets for e.g.
caldata and MAC-addresses seem to be off.
This is, howeer, not a task for the mtd NAND driver, as the flash
translation layer is tasked with this.
This patch disables this badblock shifting. This fix was originally
proposed by Jo-Philipp Wich at
https://bugs.openwrt.org/index.php?do=details&task_id=1926
Fixes FS#1926 ("MTD partition offset not correctly mapped when bad
eraseblocks present")
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 527832e54b)
That was a result of accidentally running "sed" twice on some files.
Fixes: 9b9412d55c ("treewide: replace remaining (not working now) $SAVE_CONFIG uses")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 1078de96e3)
This var has been replaced by the $UPGRADE_OPT_UPGRADE_OPT_SAVE_CONFIG
Fixes: f25d164aca ("base-files: pass "save_config" option to the "sysupgrade" method")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 5797fe84a3)
0f3c136 sysupgrade: set UPGRADE_BACKUP env variable
0bcbbbf system: fix uninitialized variables in firmware validation code
This update includes a fix for uninitialized variable usage.
Fixes: db5164d3d0 ("procd: update to the latest git HEAD")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit e8dcbbc865)
This makes sysupgrade work with the most recent procd that validates
firmware before proceeding.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit b71962da16)
This has two improvements over the current version. An autotools fix and
application of the wrt350v2 patch.
Cleaned up Makefile as a result of makefiles being fixed.
Note that this package is not really used as it depends on orion, which is
classified as broken.
This is the last package that uses svn in the tree.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit ac31ec0f62)
Ensure that the kernel objtool utilities are processed by the library
bundler in order to ensure that they're usable on foreign systems with
different libc versions.
Fixes: a9f6fceb42 ("sdk: fix building external modules when CONFIG_STACK_VALIDATION=y")
Acked-by: Yousong Zhou <yszhou4tech@gmail.com>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit efaaadb49e)
Pass suitable pkg-config overrides to the kernel build process in
order to let our pkg-config wrapper discover libraries provided
by tools/.
This mainly affects the use of libelf which is required for the
CONFIG_STACK_VALIDATION features. So far, the build system either
silently used host system libraries or kbuild simply disabled the
feature due to the lack of a suitable libelf.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit fe43969336)
When libelf from tools/ is used for building the kernel, compilation
aborts due to access to undefined defines since Kbuild adds -Wundef
to the compiler flags.
Patch the header files to use `#if defined(...)` instead of `#if ...`
to prevent such issues.
Ref: https://github.com/NixOS/nixpkgs/issues/59929
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit f3ab336d7c)
Install the pkg-config definition for libelf in order to allow the
kernel build process discover it later on.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit d3f86c9cc3)
Commit 775b70f8d5 renamed parse_file() parameters without
updating the recursive call. This broke parsing of any feeds.conf
using 'src-include'.
$ scripts/feeds update -a
Can't use string ("defaults") as a HASH ref while "strict refs" in use at scripts/feeds line 63, <$fh> line 1.
Fixes: 775b70f8d5 ("scripts/feeds: allow adding parameters to feeds")
Signed-off-by: Bjørn Mork <bjorn@mork.no>
(cherry picked from commit a21b70be31)
When CONFIG_BUILD_SUFFIX is enabled, the target-* folders in build_dir
and staging_dir have this suffix in the name, but not the
toolchain directories. When detecting the names for "arch" and "libc",
also accept the suffix and do not use it for the toolchain path.
Signed-off-by: Thomas Langer <thomas.langer@intel.com>
(cherry picked from commit 035906fd05)
Update CMake to 3.15.1
Refresh patches
Remove inofficial fossies.org and replace with GitHub (link on official site)
Remove 150-C-feature-checks-Match-warnings-more-strictly.patch as it's
a no longer needed backport from upstream.
Disable ccache if GCC is 4.8, 4.9 or 5.X to avoid build failures.
Reference: https://github.com/openwrt/openwrt/pull/1929
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
(cherry picked from commit 413c68d120)
ABI version is same.
The ipkg size increase by about 2.2%:
old:
47.909 libnftnl11_1.1.3-1_arm_cortex-a7_neon-vfpv4.ipk
new:
48.985 libnftnl11_1.1.4-1_arm_cortex-a7_neon-vfpv4.ipk
Signed-off-by: Konstantin Demin <rockdrilla@gmail.com>
(cherry picked from commit 699955a684)
This adds the key_type and ec_curve options to enable the generation of
EC keys during initialization, using openssl or the new options added to
px5g.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit 7f2b230b3b)
This adds an 'eckey' command to generate an EC key, with an optional
curve name argument, with P-256 as default.
For the 'selfsigned' command, it adds an 'ec' algorithm argument to the
'-newkey' option, and a '-pkeyopt ec_paramgen_curve:<curvename>' option,
mirroring the way openssl specifies the curve name.
Notice that curve names are not necessarily the same in mbedtls and
openssl. In particular, secp256r1 works for mbedtls, but openssl uses
prime256v1 instead. px5g uses mbedtls, but short NIST curve names P-256
and P-384 are specifically supported.
Package size increased by about 900 bytes (arm).
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit a552ababd4)
Instead, instruct the configure script to use $(FPIC) only.
Mixing -fPIC and -fpic can cause issues on some platforms like PPC.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 926157c2cc)
The configure scripts matches Linux with -fPIC, which is not exactly what
is desired. Since we are already passing $(FPIC), added a CONFIGURE_VAR to
avoid passing -fPIC.
Removed PKG_BUILD_DIR as it is already the default value.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit e2ecf39e8e)
While the -nopad option prevents mksquashfs from padding the
image to an arbitrary 4k. It does not take into consideration
that squashfs is programmed to have this 4k padding when it's
being used on on a block device... which is its main "use-case".
Now, after a week long discussion on the ML that included a
back-and-forth between some of the possible options.
But this is likely the best KISS patch to deal with the issue
right away given the limited resources.
From squashfs code point of view, be warned. The 4k padding is
not enough when dealing with devices that have a PAGE_SIZE
bigger than 4k.
if it turns out to be affecting you, then please look-up either:
"FS#2460 - kernel panic reading squashfs from ubi volume" bug
Or the discussion on the OpenWrt-Devel ML in
"amp821xx: use newly added pad-squashfs for Meraki MR24" and
"Squashfs breakage lottery with UBI..."
before making an educated guess.
Note: This will not affect the "tiny"/small flash devices as
much as it seems at first. This is because the the rootfs_data
partition that follows uses jffs2. And it requires to be aligned
to the flash block-size in order to work at all.
So either the involved FSes will meet in the middle as before,
or not at all. But in that latter case the image was already
hoping for the "undefined behaviour" gamble to turn out in its
favour and this is probably why this was unnoticed for so long.
Fixes: FS#2460
Reported-by: Russell Senior <russell@personaltelco.net>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 1c0290c5cc)
So far firmware validation result was binary limited: it was either
successful or not. That meant various limitations, e.g.:
1) Lack of proper feedback on validation problems
2) No way of marking firmware as totally broken (impossible to install)
This change introduces JSON for storing detailed validation info. It
provides a list of performed validation tests and their results. It
allows marking firmware as non-forceable (broken image that can't be
even forced to install).
Example:
{
"tests": {
"fwtool_signature": true,
"fwtool_device_match": true
},
"valid": true,
"forceable": true
}
Implementation is based on *internal* check_image bash script that:
1) Uses existing validation functions
2) Provides helpers for setting extra validation info
This allows e.g. platform_check_image() to call notify_check_broken()
when needed & prevent user from bricking a device.
Right now the new JSON info is used by /sbin/sysupgrade only. It still
doesn't make use of "forceable" as that is planned for later
development.
Further plans for this feature are:
1) Expose firmware validation using some new ubus method
2) Move validation step from /sbin/sysupgrade into "sysupgrade" ubus
method so:
a) It's possible to safely sysupgrade using ubus only
b) /sbin/sysupgrade can be more like just a CLI
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit f522047958)
this allows adding "--" prefixed parameters inside feeds.conf between the
target and name. The first parameter is --force which has the same effect
as using -f when installing any of the packages. This allows creating
feeds that will override base packages by default.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 775b70f8d5)
This explicitly lets stage2 know if partitions should be preserved. No
more "touch /tmp/sysupgrade.always.overwrite.bootdisk.partmap" hack.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit b6f4cd57e1)
9558031 system: support passing "options" to the "sysupgrade" ubus method
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 2b1a6d263c)
Update iftop to commit 77901c8c53e01359d83b8090aacfe62214658183
git log --pretty=oneline --abbrev-commit 949ed0f7..77901c8c
77901c8 Support scales beyond 1Gbps
Created with the help of the make-package-update-commit.sh script.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit cfd0748497)
957abacf Bump up version number to 1.39.2, LT revision to 32:0:18
83d362c6 Don't read too greedily
a76d0723 Add nghttp2_option_set_max_outbound_ack
db2f612a nghttpx: Fix request stall
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
(cherry picked from commit 58f929077f)
The LEDE URL is automatically redirected to the OpenWRT one,
returning an HTTP 301 code (Moved Permanently).
Also, use https, as indicated by the redirect.
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
(cherry picked from commit b39ded4ab7)
Update e2fsprogs to 1.45.3
Remove OpenBSD patch
Remove Darwin patch, neither macports or brew patches these files
Add patch to avoid crond detection on host OS
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
(cherry picked from commit 234f7a9e5d)
Convert leading spaces to tab to match rest of the file.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 7e5a0da642)
This commit will activate CONFIG_IEEE80211W for all, but the mini
variant when at least one driver supports it. This will add ieee80211w
support for the mesh variant for example.
Fixes: FS#2397
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 1d4df52c21)
Bump to latest git HEAD
509e673 firewall3: Improve ipset support
The enabled option did not work properly for ipsets, as it was not
checked on create/destroy of a set. After this commit, sets are only
created/destroyed if enabled is set to true.
Add support for reloading, or recreating, ipsets on firewall reload. By
setting "reload_set" to true, the set will be destroyed and then
re-created when the firewall is reloaded.
Add support for the counters and comment extensions. By setting
"counters" or "comment" to true, then counters or comments are added to
the set.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 51ffce0694)
Allow overriding the default selection state for Devices, similar to
setting a default for packages.
E.g. by setting DEFAULT to n, they won't be selected by default anymore
when enabling all device in the multi device profile.
This allows preventing images being built by the default config for
known broken devices, devices without enough RAM/flash, or devices not
working with a certain kernel versions.
This does not prevent the devices from being manually selected or images
being built by the ImageBuilder. These devices often still have worth
with a reduced package-set, or as a device for regression testing, when
no better device is available.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
(cherry picked from commit 7546be6007)
generate feeds.buildinfo and version.buildinfo in build dir after
containing the feed revisions (via ./scripts/feeds list -sf) as well as
the current revision of buildroot (via ./scripts/getver.sh).
With this information it should be possible to reproduce any build,
especially the release builds.
Usage would be to move feeds.buildinfo to feeds.conf and git checkout the
revision hash of version.buildinfo.
Content of feeds.buildinfo would look similar to this:
src-git routing https://git.openwrt.org/feed/routing.git^bf475d6
src-git telephony https://git.openwrt.org/feed/telephony.git^470eb8e
...
Content of version.buildinfo would look similar to this:
r10203+1-c12bd3a21b
Without the exact feed revision it is not possible to determine
installed package versions.
Also rename config.seed to config.buildinfo to follow the recommended
style of https://reproducible-builds.org/docs/recording/
Signed-off-by: Paul Spooren <mail@aparcar.org>
(cherry picked from commit 454021581f)
Force prereq again in SDK in order to fix GCC and Python dangling
symlinks:
staging_dir/host/bin/g++ -> /builder/ath79_generic/ccache_cxx.sh
staging_dir/host/bin/gcc -> /builder/ath79_generic/ccache_cc.sh
staging_dir/host/bin/python -> /usr/bin/python3.5
staging_dir/host/bin/python3 -> /usr/bin/python3.5
Ref: FS#2424
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 200f2666fb)
In the commit 623716dd43 ("comgt-ncm: Fix NCM protocol")
the dependencies to vendor NCM drivers were removed, because:
> comgt-ncm should not depend on the USB-serial-related kernel modules,
> as the cdc-wdm control device works without them. There is also no need
> to depend on kmod-huawei-cdc-ncm, since other manufacturers (like
> Ericsson and Samsung) which use other kernel modules should also be
> supported.
From a user-perspective this does not make sense, as installing comgt-ncm
(or luci-proto-ncm) should install all needed dependencies for using such
a device.
Furthermore depending on kmod-huawei-cdc-ncm does not mean that Ericsson
and Samsung devices can't be supported. By the way it seems that Ericsson
and Samsung devices never used NCM, but act as serial modems.
Thus this commit adds the dependencies again.
Signed-off-by: Vincent Wiemann <vincent.wiemann@ironai.com>
[fixed title capitalization, formatted commit message,
renamed Sony-Ericsson to Ericsson]
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit ccb4b96b8a)
8323690 state: fix shutdown when running in a container (FS#2425)
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
(cherry picked from commit d9364c1cbc)
This fixes compilation with -Werror=implicit-function-declaration.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 1b1c47577b)
GCC needs the kernel headers to compile.
Some GCC file includes asm/unistd.h which is provided by the kernel headers.
Normally the kernel headers build is very fast and ready before the gcc uses
it, but if it clones the kernel from a slow git repository it takes longer
and then it could be that the gcc already wants to use the kernel headers
before they are available. This patch fixes this problem by adding the
missing dependency.
Signed-off-by: Hauke Mehrtens <hauke.mehrtens@intel.com>
(cherry picked from commit b20156ba70)
This changes the default PKG_BUILD_DIR to take BUILD_VARIANT into
account (if set), so that packages do not need to manually override
PKG_BUILD_DIR just to handle variants.
This also updates most base packages with variants to use the updated
default PKG_BUILD_DIR.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
(cherry picked from commit e545fac8d9)
379c096 Release version 5.2.
2bce6d9 ethtool: Add 100BaseT1 and 1000BaseT1 link modes
67ffbf5 ethtool: sync ethtool-copy.h with linux-next from 30/05/2019
687152b ethtool.spec: Use standard file location macros
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
(cherry picked from commit 018395392c)
When CONFIG_ARM64_MODULE_PLTS=y, arch/arm64/kernel/module.lds is
required to build cryptodev-linux. This updates the sdk to include this
file.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
(cherry picked from commit 6e300f6a0c)
I'm trying to create a package for libgpiod, which uses
AC_CONFIG_AUX_DIR macro, which is probably leading to the following
configure error:
autoreconf: running: /openwrt.git/staging_dir/host/bin/libtoolize --force
OpenWrt-libtoolize: putting auxiliary files in AC_CONFIG_AUX_DIR, `autostuff'.
OpenWrt-libtoolize: linking file `autostuff/ltmain.sh'
OpenWrt-libtoolize: putting macros in `m4'.
...
configure: error: cannot find install-sh, install.sh, or shtool in autostuff "."/autostuff
>From the build output it's clear, that libtoolize isn't installing
install-sh symlink, because libtoolize would install install-sh only if
it's being run with --install parameter. Corresponding part in
libtoolize:
if $opt_install; then
func_config_update config.guess \
"$pkgdatadir/config" "$auxdir" pkgconfig_header
func_config_update config.sub \
"$pkgdatadir/config" "$auxdir" pkgconfig_header
func_install_update install-sh \
"$pkgdatadir/config" "$auxdir" pkgconfig_header
fi
func_ltmain_update ltmain.sh \
"$pkgdatadir/config" "$auxdir" pkgconfig_header
Adding --install parameter to libtoolize fixes this build issue:
autoreconf: running: /openwrt.git/staging_dir/host/bin/libtoolize --install --force
OpenWrt-libtoolize: putting auxiliary files in AC_CONFIG_AUX_DIR, `autostuff'.
OpenWrt-libtoolize: linking file `autostuff/config.guess'
OpenWrt-libtoolize: linking file `autostuff/config.sub'
OpenWrt-libtoolize: linking file `autostuff/install-sh'
OpenWrt-libtoolize: linking file `autostuff/ltmain.sh'
OpenWrt-libtoolize: putting macros in `m4'.
Cc: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 5cf897779e)
Making all in tests
depbase=`echo handshake_table.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\
gcc -DPACKAGE_NAME=\"libressl\" -DPACKAGE_TARNAME=\"libressl\" -DPACKAGE_VERSION=\"2.9.2\" -DPACKAGE_STRING=\"libressl\ 2.9.2\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE_URL=\"\" -DPACKAGE=\"libressl\" -DVERSION=\"2.9.2\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DLT_OBJDIR=\".libs/\" -DHAVE_SYMLINK=1 -DHAVE_ERR_H=1 -DHAVE_READPASSPHRASE_H=1 -DHAVE_ASPRINTF=1 -DHAVE_MEMMEM=1 -DHAVE_READPASSPHRASE=1 -DHAVE_STRLCAT=1 -DHAVE_STRLCPY=1 -DHAVE_STRNDUP=1 -DHAVE_STRNLEN=1 -DHAVE_STRSEP=1 -DHAVE_TIMEGM=1 -DHAVE_GETPROGNAME=1 -DHAVE_SYSLOG=1 -DHAVE_POLL=1 -DHAVE_SOCKETPAIR=1 -DHAVE_ARC4RANDOM=1 -DHAVE_ARC4RANDOM_BUF=1 -DHAVE_ARC4RANDOM_UNIFORM=1 -DHAVE_TIMINGSAFE_BCMP=1 -DHAVE_CLOCK_GETTIME=1 -DHAVE_VA_COPY=1 -DHAVE___VA_COPY=1 -DSIZEOF_TIME_T=8 -I. -I../include -I../include/compat -DLIBRESSL_INTERNAL -D__BEGIN_HIDDEN_DECLS= -D__END_HIDDEN_DECLS= -I ../crypto/modes -I ../crypto/asn1 -I ../ssl -I ../tls -I ../apps/openssl -I ../apps/openssl/compat -D_PATH_SSL_CA_FILE=\"../apps/openssl/cert.pem\" -I/Users/kevin/wrt/staging_dir/host/include -D__STRICT_ALIGNMENT -O2 -I/Users/kevin/wrt/staging_dir/host/include -fpic -Wall -std=gnu99 -fno-strict-aliasing -fno-strict-overflow -D_FORTIFY_SOURCE=2 -fstack-protector-strong -Qunused-arguments -Wno-pointer-sign -MT handshake_table.o -MD -MP -MF $depbase.Tpo -c -o handshake_table.o handshake_table.c &&\
mv -f $depbase.Tpo $depbase.Po
make[4]: *** No rule to make target `/Users/kevin/wrt/build_dir/host/libressl-2.9.2/crypto/.libs/libcrypto_la-cpuid-macosx-x86_64.o', needed by `handshake_table'. Stop.
make[3]: *** [all-recursive] Error 1
A similar error & clues from
e783d60473
"
LibreSSL 2.9.1 now has a test that requires libtls.a, however, when building a
shared library only build, the --disable-static flag is passed to libressl,
which prevents the building of libtls.a.
With libtls.a not being built, the following error occurs:
libressl-2.9.1/tls/.libs/libtls.a', needed by 'handshake_table'. Stop.
There are three options to fix this:
1) Stick with autotools, and provide a patch that removes building anything in
the tests folder.
2) Pass --enable-static to LIBRESSL_CONF_OPTS
3) Change the package type to cmake, as a cmake build does not have this issue."
It appears we cannot change to cmake because cmake has a dependency on
an ssl library.
Take option 1 and do not build the tests.
Also take the opportunity to remove man page building as well.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 8d6d227bb6)
To configure the list of allowable TLS 1.3 ciphersuites, the option
tls_ciphersuites is used instead of tls_ciphers.
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
(cherry picked from commit 261df949fa)
That function doesn't take any arguments so don't pass anything when
calling it.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 670ce65a44)
Removed an eglibc remnant, and while at it, grouped all of the
TOOLCHAIN_PLATFORMs using the same FLAGS together.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit 7d97bc19f0)
This removes package/libs/toolchain/eglibc-files/etc/nsswitch.conf.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit c47eff0df3)
do_upgrade_stage2() isn't really any common code. It isn't used anywhere
except for /sbin/sysupgrade that passes it to the stage2.
Moving its code to separated file also simplifies COMMAND variable.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit f7edd94a65)
This feature is similar to the SRC_TREE_OVERRIDE. However instead of having
to manually create a symlink inside the package folder, the buildsystem
will check if $(TOP_DIR)/src-checkout/$(PKG_NAME)/.git is present and use
this instead.
Signed-off-by: John Crispin <john@phrozen.org>
(cherry picked from commit 9e57b62783)
Update gengetopt to 2.23
Remove 100-dependency_fix.patch, fixed upstream
Rework no tests and docs patch
Fix typo in copyright section
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
(cherry picked from commit 6fac9077c6)
With bcm53xx switched to the new procedure there is no more need for
keeping that backward compatibility code.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit ea4e1dac71)
Calling nand_do_upgrade() from platform_pre_upgrade() was obsoleted in
the commit 30f61a34b4 ("base-files: always use staged sysupgrade").
This change brings bcm53xx in sync with other targets and allows
dropping backward compatibility code. It shouldn't change any user
visible behavior.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit ccf8a8e863)
Its last usage was dropped back in 2013 in the commit b95bdc8ab5
("kernel/base-files: clean up old code related to refreshing mtd
partitions, it is no longer used anywhere").
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit f58ca6ee57)
This patch adds two new commands:
zram status - shows memory stats for all zram swaps
zram compaction - trigger compaction for all zram swaps
Signed-off-by: Emil Muratov <gpm@hotplug.ru>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit a9deed62af)
No size increase on busybox binary.
Since busybox mkswap is already enabled by default it seems reasonable
to enable swapon/off too. For ex. this obsoletes installing block-mount
dependency for zram-swap.
Signed-off-by: Emil Muratov <gpm@hotplug.ru>
(cherry picked from commit afa5ce2493)
Executing '/etc/init.d/zram start' during runtime (with a swap being already
mounted) triggers zram device compaction and prints out nice stats info about
zram memory usage
Signed-off-by: Emil Muratov <gpm@hotplug.ru>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com> [use IEC's MiB unit]
(cherry picked from commit b062c90f47)
- fix dependency on BUSYBOX_CONFIG_SWAPONOFF (removed in 84da2a6)
- add busybox defaults checking (fix zram-swap always installs swap-utils
and libblkid as dependency, even if busybox includes mkswap by default)
Signed-off-by: Emil Muratov <gpm@hotplug.ru>
(cherry picked from commit c0d93432f2)
bump ABI version accordingly (thanks to Jo-Philipp Wich).
Signed-off-by: Konstantin Demin <rockdrilla@gmail.com>
(cherry picked from commit ce8027ed29)
Since we switched to 4.19, the kernel build checks for libelf to decide if
it should build tools for stack validation.
On macOS, this check fails during target/compile, but succeeds during package
build (because of the pkg-config path picking up target libraries).
Signed-off-by: Felix Fietkau <nbd@nbd.name>
(cherry picked from commit c3e31b6a9b)
All protos for wwan (ncm,qmi,mbim) do have a delay option.
To standardize that add also the missing delay option to the 3g proto.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
(cherry picked from commit 313444a79e)
Restart is in default implemented so it calls stop and start. This is
pretty unsafe to call on umount service. This service should not do
anything on restart the same way as on start. Only use of this service
is on stop.
Signed-off-by: Karel Kočí <cynerd@email.cz>
(cherry picked from commit 537b801c54)
There are mobile carrier who have different MTU size in their network.
With this change it is now possible to configure this with the qmi
proto handler.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
(cherry picked from commit 8eb63cb7df)
Having the build time written on stderr make it appear with V=w
although it is not an error or warning. Just write the time on stdout
to have it part of the build log like all the rest, but not clutter
the output when only warnings and errors should be shown.
Signed-off-by: Alban Bedel <albeu@free.fr>
(cherry picked from commit 340df72e07)
* curve25519: not all linkers support bmi2 and adx
This should allow WireGuard to build on older toolchains.
* global: switch to coarse ktime
Our prior use of fast ktime before meant that sometimes, depending on how
broken the motherboard was, we'd wind up calling into the HPET slow path. Here
we move to coarse ktime which is always super speedy. In the process we had to
fix the resolution of the clock, as well as introduce a new interface for it,
landing in 5.3. Older kernels fall back to a fast-enough mechanism based on
jiffies.
https://lore.kernel.org/lkml/tip-e3ff9c3678b4d80e22d2557b68726174578eaf52@git.kernel.org/https://lore.kernel.org/lkml/20190621203249.3909-3-Jason@zx2c4.com/
* netlink: cast struct over cb->args for type safety
This follow recent upstream changes such as:
https://lore.kernel.org/lkml/20190628144022.31376-1-Jason@zx2c4.com/
* peer: use LIST_HEAD macro
Style nit.
* receive: queue dead packets to napi queue instead of empty rx_queue
This mitigates a WARN_ON being triggered by the workqueue code. It was quite
hard to trigger, except sporadically, or reliably with a PC Engines ALIX, an
extremely slow board with an AMD LX800 that Ryan Whelan of Axatrax was kind
enough to mail me.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
(cherry picked from commit 7c23f741e9)
The linux kernel is not reproducible because the build user
and domain is included into the kernel. Set the build user
to `builder` and build domain to buildhost.
It's also possible to build reproducible builds by setting
KERNEL_BUILD_USER KERNEL_BUILD_DOMAIN to static values.
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
(cherry picked from commit fdd0a8d491)
The src-include method allows recursive inclusion of feeds.conf snippets.
This can for example be used for adding static local feeds to
feeds.conf.default without ever having to update the local feeds.conf:
src-include defaults feeds.conf.default
src-link custom /usr/local/src/lede/custom
Signed-off-by: Bjørn Mork <bjorn@mork.no>
(cherry picked from commit 7a1b575ac4)
cd7eabcd8c9d ugps: Fix compilation under 64-bit
198c06051dd0 Fix build error caused by enabled extra compiler warnings
fc2ab8756b3b Enable extra compiler warnings
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit f924fab3dc)
Declaring a package without the appropriate install section is an easy
mistake to make, especially when renaming packages. Since this is also
easy to detect, warn about it when it happens.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
No-objections-at-all-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 5ede3fcdf7)
This adds chacha20-poly1305 support to the mbedtls variant.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit 82a8ddd603)
This patch is in a series to allow additional STOP indexes after
umount, so that other block devices may stop cleanly.
rssileds.init is now STOP=89
Signed-off-by: Joseph Tingiris <joseph.tingiris@gmail.com>
(cherry picked from commit 8a5a01a677)
This patch is in a series to allow additional STOP indexes after umount,
so that other block devices may stop cleanly.
kdumpinit is now STOP=90
Signed-off-by: Joseph Tingiris <joseph.tingiris@gmail.com>
(cherry picked from commit 5883b5a1f8)
This patch is in a series to allow additional STOP indexes after umount,
so that other block devices may stop cleanly.
boot is now STOP=90
umount is now STOP=90
After this patch series, the resulting STOP indexes in the 80s & 90s
will be:
STOP=85 odhcpd.init
STOP=89 conntrackd.init
STOP=89 log.init
STOP=89 rssileds.init
STOP=90 boot
STOP=90 kdump.init
STOP=90 network
STOP=90 sysfixtime
STOP=90 umount
STOP=98 mdadm.init (note: will be addressed in a separate patch)
Signed-off-by: Joseph Tingiris <joseph.tingiris@gmail.com>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
[PKG_RELEASE is now 200]
(cherry picked from commit 04811007e5)
Packages such as xfrm contain only script files, add PKGARCH:=all
Signed-off-by: Deng Qingfang <dengqf6@mail2.sysu.edu.cn>
(cherry picked from commit 6762e72524)
'38b22b1e: deduplicate files in libnghttp2' missed duplicates in
staging_dir by Build/InstallDev.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry picked from commit ee1a783314)
The OpenWrt buildroot ABI version rebuild tracker does not handle
transient dependencies, therefor add all libraries linked by rpcd
as direct dependencies to the corresponding binary package definition.
This ensures that rpcd is automatically rebuilt and relinked if any
of these libraries has its ABI_VERSION updated in the future.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit eaad2211db)
Extend the packageauxvars database to keep a list of possible package
dependencies for each provider, then utilize this information in buildroot
to resolve the ABI version dependencies of dependent packages up to five
levels deep.
This should properly trigger rebuilds for packages indirectly depending
on other packages whose ABI_VERSION changed.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 27a4a71c24)
This is mainly for kmod-br-netfilter. To turn off
bridge-netfilter-call-xxx immediately after installation
While at it
- Define filelist="/usr/lib/opkg/info/${pkgname}.list"
- Reuse "[ -z "$root" ]"
- Grep with "-m1"
Fixes FS#2300
Reported-by: Marco Sartorius <tidbits@ormoorgmen.info>
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
(cherry picked from commit 3dc4f59eab)
It's a simple typo in the DNS file, which was pretty serious.
No scripts were working properly. Fix it up.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
[integrate upstream patch into openwrt target]
(cherry picked from commit 5470e928d0)
Support to disable the timestamp check for certificates in
wpa_supplicant (Useful for devices without RTC that cannot
reliably get the real date/time) has been accepted in the
upstream hostapd. It's implemented in wpa_supplicant as a
per-AP flag tls_disable_time_checks=[0|1].
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 99bf9a1ac2)
"There are often transient line events when
the system is powered up and initialized and it
is often necessary for the gpio_chip driver to
clear any interrupt flags in hardware before
setting up the gpio chip, especially the
irqchip portions of it."
<http://lists.infradead.org/pipermail/openwrt-devel/2019-June/017630.html>
This patch adds a fix for the APM821XX's interrupt
controller to clear any bogus pending toggled
interrupts that happens on various APM821XX boards
on boot.
The patch also changes the debouce-interval from the
default 5ms debounce interval to 60ms all around.
The default setting caused on occasions that the button
state became stuck in a pressed state, even though the
button was released.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(backported from commit 971f31bedc)
The Makefile was missing dependencies on *c_shipped, so changes never
triggered a rebuild. Add these as optional dependencies so their absence
isn't treated as an error.
In addition, fix a typo preventing the zconf.lex.o from being removed on
clean.
Fixes: 9d5510a500 ("build: add new menuconfig code based on linux 3.9")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Acked-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 0096a1cf00)
When there are multiple packages providing a meta-package, it is
possible to to create a config where a package is selected as =y, but
all of its dependency providers are just selected as =m. This is due to
the selection statement being just
config PACKAGE_foo
select PACKAGE_bar if !PACKAGE_baz
which is already fulfilled by PACKAGE_bar=m. Fix this by properly
comparing the selection states:
config PACKAGE_foo
select PACKAGE_bar if PACKAGE_baz<PACKAGE_foo
Also invert the select conditions to improve readability.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
[slightly reword commit message]
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 1fd50531ca)
After commit e82a4d9cfb ("config: regenerate *_shipped sources") the mconf
parser became more strict as a side effect and started to spew a series of
warnings when evaluating our generated kconfig sources:
tmp/.config-package.in:705:warning: ignoring unsupported character '@'
The root cause of these warnings is a wrong use of the @SYMBOL dependency
syntax in various Makefile. Fix the corresponding Makefiles by turning
`@SYM||@SYM2` expressions into the proper `@(SYM||SYM2)` form.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 74739c4228)
Since commit 31847b67bec0 ("kconfig: allow use of relations other than
(in)equality") it is possible to use relational operators in Kconfig
statements. However, those operators give unexpected results when
applied to bool/tristate values:
(n < y) = y (correct)
(m < y) = y (correct)
(n < m) = n (wrong)
This happens because relational operators process bool and tristate
symbols as strings and m sorts before n. It makes little sense to do a
lexicographical compare on bool and tristate values though.
Documentation/kbuild/kconfig-language.txt states that expression can have
a value of 'n', 'm' or 'y' (or 0, 1, 2 respectively for calculations).
Let's make it so for relational comparisons with bool/tristate
expressions as well and document them. If at least one symbol is an
actual string then the lexicographical compare works just as before.
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Tested-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
[rebased against OpenWrt kconfig, slightly reword commit message]
(backported from upstream 9059a3493efea6492451430c7e2fa0af799a2abb)
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 75dcaf3d23)
The pregenerated `zconf.hash.c` and `zconf.lex.c` files have not been
kept in sync with their respective `*.y` and `*.l` sources in the past
causing our kconfig implementation to not recognize important kconfig
grammer elements such as relational `<`, `<=`, `>` and `>=` operators.
Fixes: 2d7e602381 ("scripts/config: sync with latest linux upstream")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
[reword commit message]
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 972123f1e0)
The OpenWrt buildroot ABI version rebuild tracker does not handle
transient dependencies, therefor add all libraries linked by procd
as direct dependencies to the corresponding binary package definition.
This ensures that procd is automatically rebuilt and relinked
if any of these libraries has its ABI_VERSION updated in the
future.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 66838cd851)
7ffc239b Bump up version number to 1.39.1
bc886a0e Fix FPE with default backend
a3a14a9c Fix log-level is not set with cmd-line or configuration file
acfb3607 Update manual pages
bdfd14c2 Bump up version number to 1.39.0, LT revision to 31:4:17
cddc09fe Update AUTHORS
3c3b6ae8 Add missing colon
2f83aa9e Fix multi-line text travis issue
fc591d0c Run nghttpx integration test with cmake build
9a17c3ef travis: use multi-line text
b7220f07 cmake: Remove SPDY related files
a1556fd1 Merge pull request #1356 from nghttp2/fix-log-level-on-reload
77f1c872 nghttpx: Fix unchanged log level on configuration reload
49ce44e1 Merge pull request #1352 from nghttp2/travis-osx
f54b3ffc Fix libxml2 CFLAGS output
b0f5e5cc Implement daemon() using fork() for OSX
8d6ecd66 Enable osx build on travis
f82fb521 Update doc
2e1975dd clang-format-8
97ce392b Merge pull request #1347 from nghttp2/nghttpx-ignore-cl-te-on-upgrade
afefbda5 Ignore content-length in 200 response to CONNECT request
4fca2502 nghttpx: Ignore Content-Length and Transfer-Encoding in 1xx or 200 to CONNECT
6975c336 Update llhttp to 1.1.3
0288093c Fix llhttp_get_error_pos usage
a3a03481 Merge pull request #1340 from nghttp2/nghttpx-llhttp
c64d2573 Replace http-parser with llhttp
f028cc43 clang-format
302e3746 Merge pull request #1337 from nghttp2/upgrade-mruby
3cdbc5f5 Merge pull request #1335 from adamgolebiowski/boost-1.70
a6925186 Fix mruby build error
45d63d20 Upgrade mruby to 2.0.1
cbba1ebf asio: support boost-1.70
e86d1378 Bump up version number to 1.39.0-DEV
4a9d2005 Update manual pages
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
(cherry picked from commit 865e25e049)
Also the other type is worng and causes compile problems on ARM64
platforms.
Fixes: 9b53201d9c ("urngd: Fix wrong type in format string")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 1ae1276eab)
9dd2dcf libubox: add format string checking to ulog()
ecf5617 ustream: Add format string checks to ustream_(v)printf()
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit fc454ca153)
GCC 9.1 complains about this wrong type used in the format string, fix
this to make the compiler happy.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 9b53201d9c)
4df34a4 kmodloader: Increase path array size to make it always fit
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 22d3d91c77)
The DIR-685 has the MAC addresses in the RedBoot code just like
DNS-313. Check some magic numbers to determine that the MAC
address is where we want it and extract it from RedBoot.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
[replaced ifconfig with ip, ! -z = -n, added string quotes]
(cherry picked from commit 86098ea87e)
The DNS-313 isn't the only special board so let's bite the
bullet and create a case ladder in preparation for DIR-685.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com> [refreshed]
(cherry picked from commit 38d85d2c9f)
This simplifies the ethernet address extraction script
by using standard library functions to locate the MTD
partitions and extract ethernet address from a binary
offset location in the flash. Furthermore, the aging
ifconfig is replaced by the ip tool, which will now
assign the MAC addresses.
Suggested-by: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
[ifconfig replacement, use -n instead of ! -z,
-n requires string to be quoted within the test brackets,
drop prepended "x" in check, add quotes, make local
variables local, kill whitespaces]
(cherry picked from commit cd6c3535cd)
CONFIG_PACKET is a important feature to have.
The Packet protocol is used by applications which
communicate directly with network devices without
an intermediate network protocol implemented in
the kernel, e.g. tcpdump and ip.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit bdbb679adb)
Add proto_add_host_dependency to add a dependency to the tunlink interface
Signed-off-by: André Valentin <avalentin@marcant.net>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> [PKG_RELEASE increase]
(cherry picked from commit f6dab98044)
The f2fs-tools have a wrong PKG_LICENSE with is not SPDX compatible.
Signed-off-by: Paul Spooren <mail@aparcar.org>
(cherry picked from commit 35a70d6262)
Without this patch, an extra entry appears for AR9287 GPIO
that duplicates WLAN LED but in fact drives nothing:
gpiochip1: GPIOs 502-511, ath9k-phy0:
gpio-502 ( |netgear:blue:wlan ) out hi
gpio-503 ( |netgear:amber:test ) out hi
gpio-504 ( |netgear:green:power ) out lo
gpio-505 ( |rfkill ) in hi
gpio-507 ( |wps ) in hi
gpio-508 ( |reset ) in hi
gpio-510 ( |ath9k-phy0 ) out hi <===!
The pin pointed above is default LED GPIO (8) for AR9287.
For WNR2200 it is not connected anywhere - pin 0 drives blue WLAN
LED instead - but initialization code is missing that information.
This fix calls ap9x_pci_setup_wmac_led_pin() function at device
setup, forcing WLAN LED pin to be 0 and removing redundant entry.
Signed-off-by: Michal Cieslakiewicz <michal.cieslakiewicz@wp.pl>
The button is labelled reboot/restore in documentation, and has always
been used for that. Naming it WPS has always been wrong.
Signed-off-by: Karl Pálsson <karlp@etactica.com>
[matched author with SoB]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
The command line enforced by the kernel is harmful, we want to
use the command line passed from the device tree. This is
especially important for the SQ201, Teltonika RUT and WBD111
which has a console set to 115200n8 and not 19200n8 like all
the others, so their consoles will be confusingly changing
baudrate between boot loader and kernel. The setting in the
device tree corresponds to what the boot loader is using.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit ea44ed5f38)
Preserve optionality of libcap by having configuration script follow the
HAVE_CAP environment variable, used similarly to the HAVE_ELF variable.
Signed-off-by: Alin Nastac <alin.nastac@gmail.com>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> [PKG_RELEASE increase/refresh patches]
(backported from commit a6da3f9ef7)
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
By default, set BE tx queue TXOP limit to 2.0 in the hostapd config
Many vendor drivers are doing similar things to boost throughput.
On MT7612 under ideal conditions, it improves tx throughput from 470 Mbit/s
to about 570 Mbit/s.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
(cherry-picked from commit 8650201f10)
Close cooperation with Lorenzo Bianconi resulted
in these patches which fix all remaining seen issues
when using dynack.
Fix link losses when:
- Late Ack's are not seen or not present
- switching from too low static coverage class to dynack on a live link
These are fixed by setting the Ack Timeout/Slottime to
the max possible value for the currently used channel width when
a new station has been discovered.
When traffic flows, dynack is able to adjust to optimal values
within a few packets received (typically < 1 second)
These changes have been thoroughly tested on ~60 offshore devices
all interconnected using mesh over IBSS and dynack enabled on all.
Distances between devices varied from <100m up to ~35km
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
(cherry picked from commit f6e8ba0238fe349b7529357793e2fb18635819ed)
6a61b9a probe: fallback to libblkid.so.1 when libblkid.so does not exist
Also remove deprecation notices from init script while we're at it.
Fixes: FS#2274
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 517cb0b70b)
Update fstools to commit 1539b535ac327a3bc599d1ca871e14fd0dc3bba1
git log --pretty=oneline --abbrev-commit ff1ded63..1539b535
1539b53 libblkid-tiny: increment label size to 256
d563f3c libblkid-tiny: fix wrong btrfs label length
3957dd3 block: prevent mount point confusion
9b36dc2 libfstools: avoid false positives when matching devices and volumes
Created with the help of the make-package-update-commit.sh script.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 8036345225)
a9f9557 nl80211: support reading hardware id from phy directly
c586cd3 iwinfo: add device id for MediaTek MT7612E
d4382dd iwinfo: add device id for Atheros AR9390
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit b13f3300d5)
tx_size was just declared above and set to BIT(tx->order)
Use the declaration instead, which could avoid a pointer deref
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 6852eab58236e3991f5d73f211018dfef01f2a64)
Backport of a4eef43a12 ("ath79: ag71xx: replace alloc_etherdev with devm_alloc_etherdev")
combined with the initial changes from John Crispin.
Simplifies the code a lot by using the Managed dev API.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Backport of 4eaa3626a8 ("ath79: ag71xx: pass correct device pointer to dma functions")
While 4.14 does not contain the warnings,
it still makes sense to use the proper pointers here.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
All other instances of this identical declaration fetch the
value directly from the ring_order.
Also do it here.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Removed upstream patch
Compile and run tested on mvebu
Signed-off-by: DENG Qingfang <dengqf6@mail2.sysu.edu.cn>
(cherry picked from commit fb57c617ba5269a6c2050e42b9d68813e262bf0d)
Refreshed all patches.
Also add a missing symbol for x86 which got used now in this bump.
- ISCSI_IBFT
Compile-tested on: cns3xxx, x86_64
Runtime-tested on: cns3xxx, x86_64
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
This should fix a problem with 1560 MTU, 160Mhz on DFS channels,
some other small issues on < 5.2 kernels, and for 5.2 driver,
it pulls in some upstream stable fixes.
wave-1 firmware changes since last update:
* June 24, 2019: Try allocating low-priority WMI msgs if high-prio are not available.
* June 24, 2019: Init rate-ctrl to start at lowest rate instead of in the middle. Hoping
this helps DHCP when station connects from a long distance.
wave-2:
* June 24, 2019 Start rate-ctrl at minimal values to help DHCP work better for far-away peers.
* July 24, 2019 Fix old regression that made /a (and probably /b/g) perform poorly, at least on
diet-compiled images.
* Aug 8, 2019 Improve a/b/g rate-ctrl by damping the PER swings caused by the all-or-nothing logic
of transmitting non-block-ack frames one at a time.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit de5f0764883ad6a0767da58d7359f3e01aa91139)
the ath10k-ct package ships multiple versions of the ath10k-ct driver,
OpenWrt currently only uses the version 4.19, but we still ship some
patches for older versions. Remove all patches only touching older
versions and also remove the patch for older versions from patches which
do the same changes to multiple versions of ath10k-ct.
This removes some unneeded patches, the end binary should stay the same.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 10fe5ca362)
This patch updates the board-2.bin for the default
IPQ4019, QCA9984 and QCA9888 ath10k-firmware-xyz-ct
and -ct-htt firmwares.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 7f79882d44)
wave-1:
2019-05-09: Tweak rate-ctrl: Ramp PER up faster, down slower. This
helps throughput in rate-vs-range test, especially with
nss1.
2019-05-20: Disable adaptive-CCA. I am not sure it helps, and it may
make it slower to detect noise that should tell the system
to stop transmitting. If someone has means to test this
properly, I'd be happy to work with them.
wave-2:
2019-05-15: Fix problem where rate-ctrl sometimes used rix of 0x0.
2019-05-15: Allow raw-tx of encrypted frame. Requires a patch to the
driver to use raw mode when skb has WEP flag enabled AND
skb is flagged to not be encrypted. Lightly tested.
2019-05-16: Fix tx-hang that happened when rate-ctrl chose an OFDM rate
for 20Mhz and sent that as AMPDU. To fix, limit to (V)HT
rates if peer is (V)HT. It seems that MCS0 (V)HT20 should
have as good of a chance of being detected as CCK or OFDM.
2019-06-06: Disable TX-BFEE, TX-BFER for IBSS connections. I suspect
this is part of the tx-hang issue seen with IBSS between
two 9984 radios.
2019-06-12: Fix rx-rate reporting in 'fw_stats' logic. This was at
least partly due to regressions I had added earlier when
working on some multi-vdev enhancements.
2019-6-12: Fix case where extd peer-stats were not always populated.
The stats gathering code did not handle error conditions
well.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit ff2382e36c)
Changes:
ath10k: Improve PMF/MPF mgt frame check
And add a driver for 5.2 (beta, not even tested yet) kernel.
Refresh patches.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 49b3dcb2ab)
For x86/64 (maybe more) target the SUPPORTED_DEVICES variable is empty
which causes the `&&` junction to fail, producing a non zero exit code.
Tested-by: Paul Spooren <mail@aparcar.org>
Fixed-by: Jo-Philipp Wich <jo@mein.io>
Signed-off-by: Paul Spooren <mail@aparcar.org>
(cherry picked from commit 07926d7def)
This adds missing includes for sys/sysmacros.h which are needed with
musl libc 1.1.23.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit ba8aeb02ea)
musl libc through 1.1.23 has an x87 floating-point stack adjustment
imbalance, related to the math/i386/ directory. In some cases, use of
this library could introduce out-of-bounds writes that are not present
in an application's source code.
This problem only affects x86 and no other architectures.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit c262daf308)
The OpenWrt buildroot ABI version rebuild tracker does not handle
transient dependencies, therefor add all libraries linked by uhttpd
as direct dependencies to the corresponding binary package definition.
This ensures that uhttpd is automatically rebuilt and relinked if any
of these libraries has its ABI_VERSION updated in the future.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit a95ddaba02)
This happens only the second time a library is loaded by dlopen().
After lib1 is loaded, dlsym(lib1,"undef1") correctly resolves the undef
symbol from lib1 dependencies. After the second library is loaded,
dlsym(lib2,"undef1") was returning the address of "undef1" in lib2
instead of searching lib2 dependencies.
Using upstream fix which now uses the same logic for relocation time
and dlsym.
Fixesopenwrt/packages#9297
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
(cherry picked from commit 0d0617ff14)
It seems bzip2 was abandoned by the author and adopted by the sourceware
people. The last release of bzip2 was from 2010.
Several security bugs were fixed as well as others.
Fixed up PKG_LICENSE to be compatible with SPDX.
Changed URLs to point to the new home.
Added patch that gets rid of deprecated utime function and switches it to
utimensat.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Always build AES-GCM support.
Unnecessary patches were removed.
This includes two vulnerability fixes:
CVE-2019-11873: a potential buffer overflow case with the TLSv1.3 PSK
extension parsing.
CVE-2019-13628 (currently assigned-only): potential leak of nonce sizes
when performing ECDSA signing operations. The leak is considered to be
difficult to exploit but it could potentially be used maliciously to
perform a lattice based timing attack.
This brings the package up-to-date with master, so it incorporates
changes from 4.0.0 in master:
* Removed options that can't be turned off because we're building with
--enable-stunnel, some of which affect hostapd's Config.in.
* Adjusted the title of OCSP option, as OCSP itself can't be turned off,
only the stapling part is selectable.
* Mark options turned on when wpad support is selected.
* Add building options for TLS 1.0, and TLS 1.3.
* Add hardware crypto support, which due to a bug, only works when CCM
support is turned off.
* Reorganized option conditionals in Makefile.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
This updates to backports-4.19.57-1 which contains the wireless
subsystem and driver from kernel 4.19.57.
The removed patches are applied upstream.
Tested-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This will reduce the size of the kernel if CONFIG_CC_OPTIMIZE_FOR_SIZE is
set like for all targets with small_flash feature flag.
I haven't seen any changes for an ARM64 target which optimizes the
kernel for speed instead.
On the ath79/tiny target the uncompressed kernel size was reduced by
3.2% and the compressed kernel size by 2.1%
kernel size with CONFIG_OPTIMIZE_INLINING=n
4346412 build_dir/target-mips_24kc_musl/linux-ath79_tiny/vmlinux
1391169 build_dir/target-mips_24kc_musl/linux-ath79_tiny/tplink_tl-wr941-v4-kernel.bin
Kernel size with CONFIG_OPTIMIZE_INLINING=y
4212396 build_dir/target-mips_24kc_musl/linux-ath79_tiny/vmlinux
1362051 build_dir/target-mips_24kc_musl/linux-ath79_tiny/tplink_tl-wr941-v4-kernel.bin
This change is currently pending for kernel 5.2 and already in
linux-next, this updates our patch to match the upstream version.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 6dac1c0a9b)
Unconditionally execute the final case statement, even if the disk layout
changed. This is necessary, to keep the original Turris Omnia flash
instructions working: The disk layout WILL change, when switching from
TurrisOS to OpenWRT. Without updating the uboot environment at the same
time, the user would end up with an unbootable system.
Fixes commit 2e5a0b81ec ("mvebu: sysupgrade: sdcard: keep user added ...")
Signed-off-by: Klaus Kudielka <klaus.kudielka@gmail.com>
This commit fixes regression on Linksys WRT1900 (Mamba) where this device
doesn't have USB 3.0 controller integrated in SoC, instead it has Etron
EJ168 connected to PCIe lane. Previously enabled in kernel 4.4 and 4.9,
was lost in transition to 4.14.
Fixes: 4ccad92 ("mvebu: Add support for kernel 4.14")
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
This commit unifies the LED mapping of the AVM Fritz!Box routers, which
have a combined Power/DSL LED.
With the stock firmware, the Power LED has the following
characteristics:
- Blink when DSL sync is being established
- Solid when DSL sync is present
We can't completely resemble this behavior in OpenWrt. Currently, the
Power LED is completely off, when DSL sync is missing. This is not
really helpful, as a user might have the impression, that he bricked his
device.
Instead, map the Info-LED to the state of the DSL connection.
There is no consistent behavior for the Info-LED in the stock
firmware, as the user can set it's function by himself. The DSL
connection state is one possible option for the Info LED there.
Also use the red Power LED to indicate a running upgrade, in case the
board has a two-color Power LED.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 3032bf7f89)
This reverts commit 2b074654b0.
Due to popular demand, I've decided to revert this commit after
deliberating with Petr. This means that 19.07 will ship ath79 binary
images after all.
We do however consider the ath79 target on Kernel 4.14 in the 19.07
branch to be beta quality at best.
Acked-by: Petr Štetiar <ynezz@true.cz>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
GNU patch through 2.7.6 is vulnerable to OS shell command injection that
can be exploited by opening a crafted patch file that contains an ed style
diff payload with shell metacharacters. The ed editor does not need to be
present on the vulnerable system. This is different from CVE-2018-1000156.
https://nvd.nist.gov/vuln/detail/CVE-2019-13638
Signed-off-by: Russell Senior <russell@personaltelco.net>
(cherry picked from commit bcfd1d7685)
This adds the led-upgrade alias for the AVM Fritz!Box 7412 to indicate a
running firmware upgrade.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 920abb9b58)
The AVM Fritz!Box 7412 does not use the VMMC part of the Lantiq chip but
rather a proprietary solution based on the DECT chip for the FXS ports.
Therefore, the second VPE can be enabled for use with OpenWrt.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 891a700759)
The AVM FRITZ!Box 7412 buttons are both active low, which is currently
incorrectly defined in the device-tree.
This leads to the device booting directly into failsafe.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit c12947b39e)
This commit selects wpad-basic for the FRITZ!Box 7312 and 7412 as
wpad-mini is only selected on boards with small flash.
Signed-off-by: Johann Neuhauser <johann@it-neuhauser.de>
[add short description]
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit c3c3cd5e4a)
ar71xx got lost during final rebase ..
Fixes: c8a6ce71e4 ("ar71xx/ath79: ag71xx: init rings with GFP_KERNEL")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Upstream commit 246902bdf562d45ea3475fac64c93048a7a39f01
Which contains following explanation:
--
There is no need to use GFP_ATOMIC here, GFP_KERNEL should be enough.
The 'kcalloc()' just a few lines above, already uses GFP_KERNEL.
--
Looking at the code, all other descriptors also use plain GFP_KERNEL
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Introduce a new option CONFIG_SIGNATURE_CHECK which defaults to the value
of CONFIG_SIGNED_PACKAGES and thus is enabled by default.
This option is needed to support building target opkg with enabled
signature verification while having the signed package lists disabled.
Our buildbots currently disable package signing globally in the
buildroot and SDK to avoid the need to ship private signing keys to
the build workers and to prevent the triggering of random key generation
on the worker nodes since package signing happens off-line on the master
nodes.
As unintended side-effect, updated opkg packages will get built with
disabled signature verification, hence the need for a new override option.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit f565f276e2)
Fix build breakage as upstream has removed implicit include of
sys/sysmacros.h from sys/types.h:
remove implicit include of sys/sysmacros.h from sys/types.h
this reverts commit f552c792c7ce5a560f214e1104d93ee5b0833967, which
exposed the sysmacros.h macros (device major/minor calculations) for
BSD and GNU profiles to mimic an unintentional glibc behavior some
code depended on. glibc has deprecated and since removed them as the
resolution to bug #19239, so it makes no sense for us to keep this
behavior. affected code should all have been fixed by now, and if it's
not yet fixed it needs to be for use with modern glibc anyway.
Ref: https://git.musl-libc.org/cgit/musl/commit/include/sys/types.h?id=a31a30a0076c284133c0f4dfa32b8b37883ac930
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 79596f782e)
Since usign miscalculates SHA-512 digests for input sizes of exactly
64 + N * 128 + 110 or 64 + N * 128 + 111 bytes, we need to apply some
white space padding to avoid triggering the hashing edge case.
While usign itself has been fixed already, there is still many firmwares
in the wild which use broken usign versions to verify current package
indexes so we'll need to carry this workaround in the forseeable future.
Ref: https://forum.openwrt.org/t/signature-check-failed/41945
Ref: https://git.openwrt.org/5a52b379902471cef495687547c7b568142f66d2
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit e1f588e446)
This update fixes usign signature verification on files with certain
file sizes triggering a bug in the shipped SHA-512 implementation.
5a52b37 sha512: fix bad hardcoded constant in sha512_final()
3e6648b README: replace unicode character
716c3f2 README: add reference to OpenBSD signify
86d3668 README: provide reference for ed25519 algorithm
939ec35 usign: main.c: describe necessary arguments for -G
Ref: https://forum.openwrt.org/t/signature-check-failed/41945
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 991dd5a893)
WLAN0 and the unused LED are currently swapped. Fix this, so the LED
behavior matches the other OCEDo devices.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit a3a2b4857e)
New Docker based buildslaves install just bare minimum of packages, thus
not having bsdmainutils package installed which provides `hexdump`
utility, leading to the following build breakage on buildbots:
ubinize-image.sh: 12: /builder/scripts/ubinize-image.sh: hexdump: not found
So this patch simply replaces `hexdump` with `od` utility provided by
coreutils package, which should be likely available.
Co-authored-by: Yousong Zhou <yszhou4tech@gmail.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit c6d41c320c)
The Aerohive HiveAP 121 has the wrong PLL value set for Gigabit speeds,
leading to packet-loss. 10M and 100M work fine.
This commit sets the Gigabit Ethernet PLL value to the correct value,
fixing packet loss.
Confirmed with iperf and floodping.
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit cb49e46a8a)
commit e09da0169a ("ar71xx: fix Mikrotik board detection")
was generated based on testing a rb-912 board, on which detection failed.
Testing on more hardware shows something fun:
machine : MikroTik RouterBOARD 922UAGS-5HPacD
machine : Mikrotik RouterBOARD 912UAG-5HPnD
Both lowercase and uppercase are used.
So ensure we support both now ..
Fixes: e09da0169a ("ar71xx: fix Mikrotik board detection")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 845b2a1cfe)
new features:
- riscv64 port
- configure now allows customizing AR and RANLIB vars
- header-level support for new linux features in 5.1
major internal changes:
- removed extern __syscall; syscall header code is now fully self-contained
performance:
- new math library implementation for log/exp/pow
- aarch64 dynamic tlsdesc function is streamlined
compatibility & conformance:
- O_TTY_INIT is now defined
- sys/types.h no longer pollutes namespace with sys/sysmacros.h in any profile
- powerpc asm is now compatible with clang internal assembler
changes for new POSIX interpretations:
- fgetwc now sets stream error indicator on encoding errors
- fmemopen no longer rejects 0 size
bugs fixed:
- static TLS for shared libraries was allocated wrong on "Variant I" archs
- crash in dladdr reading through uninitialized pointer on non-match
- sigaltstack wrongly errored out on invalid ss_size when doing SS_DISABLE
- getdents function misbehaved with buffer length larger than INT_MAX
- set*id could deadlock after fork from multithreaded process
arch-specfic bugs fixed:
- s390x SO_PEERSEC definition was wrong
- passing of 64-bit syscall arguments was broken on microblaze
- posix_fadvise was broken on mips due to missing 7-arg syscall support
- vrregset_t layout and member naming was wrong on powerpc64
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
- add uart rom script address in header of sdma firmware to support
the uart driver of latest kernel working well while old firmware
assume ram script used for uart driver as NXP internal legacy
kernel.
- add multi-fifo SAI/PDM scripts.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 819b6345a206ad182dd3c2d786a3d7f04e33f751)
In order to prevent build of images for this release as ath79 is going
to deprecate ar71xx in the next release.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Fix a typo in the machine type being extracted from /proc/cpuinfo
which causes all Mikrotik board to be undetected properly.
This lead to sysupgrade issues and probably some others too.
Fixes: 76c963bb01 ("ar71xx: base-files: fix board detect on new MikroTik devices")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Update WLAN LED colour identifier for both interfaces on Archer C7
Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>
(cherry picked from commit 65762cdd22)
Move system LED board definitions of Archer C5/C7 to reflect
actual system LED colour used
Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>
(cherry picked from commit a73934fc9a)
All leds on these boards are green. v1 has RFKILL GPIO 23 for production
units (it had GPIO 13 only for test phase units, and these are rather
very rare to find). As for the previous attempt to fix this and revert
due to WDR boards have blue leds, it was wrong: WDR board does not use
common setup (false).
Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>
(cherry picked from commit c79c001b59)
Base address for USB0 has changed from 0x18116c94 on AR934X
to 0x18116d94 on QCA9558. CP Typo remained for years here...
Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>
(cherry picked from commit fd5aa19480)
- make create_pbl and byte_swap as host tools
- fix a bug that maybe use the cross compiler
to compile create_pbl and byte_swap:
# -a option appends the image for Chassis 3 devices in case of non secure boot
aarch64-openwrt-linux-musl-gcc -Wall -Werror -pedantic -std=c99 -O2
-DVERSION=v1.5(release):reboot-10604-ge9216b3336 -D_GNU_SOURCE -D_XOPEN_SOURCE=700
-c -o create_pbl.o create_pbl.c
cc1: note: someone does not honour COPTS correctly, passed 0 times
LD create_pbl
/usr/bin/ld: create_pbl.o: Relocations in generic ELF (EM: 183)
/usr/bin/ld: create_pbl.o: Relocations in generic ELF (EM: 183)
/usr/bin/ld: create_pbl.o: Relocations in generic ELF (EM: 183)
create_pbl.o: error adding symbols: File in wrong format
collect2: error: ld returned 1 exit status
Makefile:43: recipe for target create_pbl failed
make[4]: *** [create_pbl] Error 1
plat/nxp/tools/pbl_ch2.mk:45: recipe for target pbl failed
make[3]: *** [pbl] Error 2
- add tfa- prefix to all tools in order to avoid future clashes with
other toolnames
Signed-off-by: Biwen Li <biwen.li@nxp.com>
[added missing HOST_CFLAGS, added tfa- prefix to the tools]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 83d5ca2186)
fiptool is a host tool, used in a firmware generation pipeline, but it's
not treated as such, leading to the build breakage on the hosts which
don't have {Open,Libre}SSL dev package installed:
In file included from fiptool.h:16:0,
from fiptool.c:19:
fiptool_platform.h:18:27: fatal error: openssl/sha.h:
No such file or directory
# include <openssl/sha.h>
So this patch promotes fiptool into the host tool with proper host
include and library paths under STAGING_DIR.
Ref: https://github.com/openwrt/openwrt/pull/2267
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit b8249cef9f)
urandom-seed content was split from base-files into separate package so
in order to preserve the current functionality and to provide some
fallback mechanism in case jent-rng initialization fails in urngd we
need to add it back.
urngd is OpenWrt's micro non-physical true random number generator based
on timing jitter.
Tested-by: Lucian Cristian <lucian.cristian@gmail.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit f5387b754f)
μrngd is OpenWrt's micro non-physical true random number generator based
on timing jitter.
Using the Jitter RNG core, the rngd provides an entropy source that
feeds into the Linux /dev/random device if its entropy runs low. It
updates the /dev/random entropy estimator such that the newly provided
entropy unblocks /dev/random.
The seeding of /dev/random also ensures that /dev/urandom benefits from
entropy. Especially during boot time, when the entropy of Linux is low,
the Jitter RNGd provides a source of sufficient entropy.
Acked-by: Jo-Philip Wich <jow@mein.io>
Tested-by: Lucian Cristian <lucian.cristian@gmail.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 714bd89fce)
This doesn't include 9ff8614a3dbe ("brcmfmac: use separate Kconfig file
for brcmfmac") due to a few conflicts with backports changes.
An important change is:
[PATCH 2/7] brcmfmac: change the order of things in brcmf_detach()
which fixes a rmmod crash in the brcmf_txfinalize().
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit db8e08a5a4)
SoC: MediaTek MT7621AT
RAM: 128M (Winbond W631GG6KB-15)
FLASH: 16MB (Spansion S25FL128SA)
WiFi: MediaTek MT7603EN bgn 2SS
WiFi: MediaTek MT7612EN nac 2SS
BTN: Reset - WPS
LED: - Power
- LAN {1-4}
- WAN
- WiFi 2.4 GHz
- WiFi 5 GHz
- USB
UART: UART is present next to the Power LED.
TX - RX - GND - 3V3 / 57600-8N1
3V3 is the nearest one to the Power LED.
Installation
------------
Via TFTP:
1. Set your computers IP-Address to 192.168.1.75.
2. Power up the Router with the Reset button pressed.
3. Release the Reset button after 5 seconds.
4. Upload OpenWRT sysupgrade image via TFTP:
> tftp -4 -v -m binary 192.168.1.1 -c put <IMAGE>
Via SSH:
Note: User/password for SSH is identical with the one used in the
Web-interface.
1. Complete the initial setup wizard.
2. Activate SSH under "Administration" -> "System".
3. Transfer the OpenWrt sysupgrade image via scp:
> scp owrt.bin admin@192.168.1.1:/tmp
4. Connect via SSH to the router.
> ssh admin@192.168.1.1
5. Write the OpenWrt image to flash.
> mtd-write -i /tmp/owrt.bin -d linux
6. Reboot the router
> reboot
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 14e0e4f138)
Hardware
--------
CPU: Qualcomm Atheros QCA9558
RAM: 128M DDR2
FLASH: 16MiB
ETH: 1x Atheros AR8035 (PoE in)
WiFi2: QCA9558 3T3R
WiFi5: QCA9880 3T3R
BTN: 1x Reset
LED: 1x LED Power (non-controllable)
1x LED Status (internal)
1x LED LAN (controlled by PHY)
1x LED WLAN
BEEP: 1x GPIO attached piezo beeper
UART: 3.3V GND TX RX (115200-N-8) (3.3V is square pad)
Header is located next to external-LED header.
Installation
------------
Make sure you set a password for the root user as prompted on first
setup!
1. Upload OpenWRT sysupgrade image via SSH to the device.
Use /tmp as the destination folder on the device.
User is root, password the one set in the web interface.
2. Install OpenWRT with
> sysupgrade -n -F /tmp/<openwrt-image-name>
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit cb3cd52694)
fixes intermittent loss of connectivity on 1Gbit port, with log message:
> 803x_aneg_done: SGMII link is not ok
Thanks to David Bauer for pointing me in the right direction.
I just had to figure out the right bus_id, which you find in this log:
> ag71xx ag71xx.1: connected to PHY at gpio-1:00 [uid=004dd074,
driver=Atheros 8031 ethernet]
Fixes FS#2236
Signed-off-by: Etienne Champetier <champetier.etienne@gmail.com>
[Wrapped commit message - Fixed whitespace erors]
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 2a7519e29d)
Image generation is currently failing on builbots due to the following
error:
WARNING: Image file [...] mt7620-ex2700-squashfs-factory.bin is too big
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Many Atom-based embedded/industrial x86 boards can't run 64bit operating
systems due to either processor or board firmware limitations, but they
have modern interfaces (PCIe) or have modern Intel gigabit controllers
onboard. With the current default package selection for x86 Generic
target their network won't work.
Add the modern gigabit network modules needed or most likely going to be
used as add-in cards, similar to what is the list on x86_64 target.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
[fixed whitespace issue]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit bb27cde257)
Image generations is currently failing on builbots due to the following
errors:
mkfwimage -B XS2 -v XS2 [...] ath25-ubnt2-squashfs-sysupgrade.bin
ERROR: Failed creating firmware layout description - error code: -2
mkfwimage -B XS5 -v XS5 [...] ath25-ubnt5-squashfs-sysupgrade.bin
ERROR: Failed creating firmware layout description - error code: -2
That cryptic -2 error simply means, that kernel+rootfs doesn't fit into
the firmware partition.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
This patch unifies the polled and interrupt-driven gpio_keys code
paths as well implements consistent handling of the debounce
interval set for the GPIO buttons and switches.
Hotplug events will only be fired if
1. The input changes its state and remains stable for the duration
of the debounce interval (default is 5 ms).
2. In the initial stable (no state-change for duration of the
debounce interval) state once the driver module gets loaded.
Switch type inputs will always report their stable state.
Unpressed buttons will not trigger an event for the initial
stable state. Whereas pressed buttons will trigger an event.
This is consistent with upstream's gpio-key driver that uses
the input subsystem (and dont use autorepeat).
Prior to this patch, this was handled inconsistently for interrupt-based
an polled gpio-keys. Hence this patch unifies the shared logic into the
gpio_keys_handle_button() function and modify both implementations to
handle the initial state properly.
The changes described in 2. ) . can have an impact on the
failsafe trigger. Up until now, the script checked for button
state changes. On the down side, this allowed to trigger the
failsafe by releasing a held button at the right time. On the
plus side, the button's polarity setting didn't matter.
Now, the failsafe will only engage when a button was pressed
at the right moment (same as before), but now it can
theoretically also trigger when the button was pressed the
whole time the kernel booted and well into the fast-blinking
preinit phase. However, the chances that this can happen are
really small. This is because the gpio-button module is usually
up and ready even before the preinit state is entered. So, the
initial pressed button event gets lost and most devices behave
as before.
Bisectors: If this patch causes a device to permanently go into
failsafe or experience weird behavior due to inputs, please
check the following:
- the GPIO polarity setting for the button
- the software-debounce value
Run-tested for 'gpio-keys' and 'gpio-keys-polled' on
- devolo WiFi pro 1200e
- devolo WiFi pro 1750c
- devolo WiFi pro 1750x
- Netgear WNDR4700
- Meraki MR24
- RT-AC58U
Signed-off-by: David Bauer <mail@david-bauer.net>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com> [further
cleanups, simplification and unification]
(cherry picked from commit 27f3f493de)
While testing 4.19 build on malta/be64, I've encountered following
error:
gpio-button-hotplug/gpio-button-hotplug.c:529:18: error: implicit
declaration of function 'gpio_to_desc'
which is caused by the missing include fixed by this patch.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit dd6d82112a)
In ath79, identifiers tplink_tl-wdr3600 and tplink_tl-wdr4300 have
been used while most other TP-Link devices include the revision.
Although there actually is only one major revision of these
devices, they bear the revision on their bottom (v1.x). TP-Link
also refers to the devices as V1 on its web page.
This patch thus adds -v1 to both so it is more consistent
with other devices and with what you would expect from reading
the on-device sticker and the support pages.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
This addresses an issue where the list option specified in
/etc/config/openvpn i.e. 'tls_cipher' would instead show up in the
generated openvpn-<name>.conf as 'ncp-ciphers'. For context,
'ncp_ciphers' appears after 'tls_cipher' in OPENVPN_LIST from
openvpn.options.
Also, the ordering of the options in the UCI config file is now
preserved when generating the OpenVPN config. The two currently
supported list options deal with cipher preferences.
Signed-off-by: Matt Merhar <mattmerhar@protonmail.com>
(cherry picked from commit 1d4c4cbd20)
It was reported in FS#2385, that Carambola2 doesn't currently have
working watchdog so fix it by adding watchdog node.
Ref: FS#2385
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 0893f28e19)
The OpenWrt buildroot ABI version rebuild tracker does not handle
transient dependencies, therefore add all libraries linked by
block-mount and blockd as direct dependencies to the corresponding
binary package definition.
This ensures that block-mount and blockd is automatically rebuilt and
relinked if any of these libraries has its ABI_VERSION updated in the
future.
Fixes: FS#2373
[jow: similar fix for procd and 98.42% of commit message]
Signed-off-by: Jo-Philip Wich <jow@mein.io>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit cbae306815)
Now that busybox is a known alternatives provider by opkg, we remove the
ALTERNATIVES spec and add a note to make the implicit situation clear
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
(cherry-picked from commit 62be427067)
Opkg starting from this version special-cases busybox as alternatives
provider. There should be no need to add entries to ALTERNATIVES of
busybox package
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
(cherry-picked from commit e51b513f75)
While flashing lots of RB2011 devices, I noticed that
some of them refused to boot properly, failing over the NAND parameters.
Checking in detail shows that some device seem to use another NAND flash
which only support standard 2048-byte pages, without 512-byte subpage support.
This commit disables usage of these small subpage completely.
Advantages:
- Both NAND's with(out) subpage support are working now
- The nand speed increases a bit (measured roughly 1%) in typical usecases
Disadvantages:
- The maximum storage capacity decreases by ~0.2%
as small changes can consume a full page (2048 bytes) now.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Commit 000d400baa ("kernel: drop everything not on kernel version 4.14")
dropped everything not related to kernel 4.14, but forgot
to remove the parts in kernel-version.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
b43legacy needs ssb support and we do not compile the mips74 subtarget
of the brcm47xx target with SSB support. This causes a build failure in
the mac80211 package and only some of the kernel modules are being
created.
I am not aware of any device with a BRCM47xx mips74 CPU which uses a
b43legacy compatible device.
Fixes: FS#2334
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit e05310b9b8)
This is already included in newer upstream. Needed to build BPF programs
using the MIPS kernel include files.
Without this patch, clang fails with "#error Use a Linux compiler or
give up." in sgidefs.h when building BPF programs.
Signed-off-by: Fredrik Olofsson <fredrik.olofsson@anyfinetworks.com>
(cherry-picked from commit 7d96c301d6)
This includes a fix for a medium-level potential cache attack with a
variant of Bleichenbacher’s attack. Patches were refreshed.
Increased FP_MAX_BITS to allow 4096-bit RSA keys.
Fixed poly1305 build option, and some Makefile updates.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit 2792daab5a)
Currently, path argument is only checked for being not empty.
This changes behavior to actually check whether path exists.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
After commit 1e41de2f48 ("mpc85xx: convert TL-WDR4900 v1 to simpleImage")
XZ compression of zImage was enabled. This change exposed a problem with
the HiveAP-330 images, which was fixed by foregoing the compression on
the kernel altogether with commit 98089bb8ba
("mpc85xx: Use uncompressed kernel on the HiveAP-330").
This patch adds back the gzip compression of the kernel image by
utilizing the generic OpenWRT uImage method instead of relying on
the PowerPC bootwrapper script that did it previously.
Compile-tested: p1020/hiveap-330
Tested-by: Chris Blake <chrisrblake93@gmail.com> [run-tested]
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
[filled in even more text]
While moving common used parts to dtsi files, the was disabled by
default but not enabled for all boards using the STP.
Fixes: f519fea4c6 ("lantiq: kernel 4.14: cleanup dts files")
Reported-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Mathias Kresin <dev@kresin.me>
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Apply the same approach as in commit 3b53d6fdbc ("ar71xx: fix pci irq
init on kernel 4.14") to fix IRQ initialization for ath79-based chipsets
on rb4xx.
Ref: PR#2182
Acked-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Signed-off-by: W. Michael Petullo <mike@flyn.org>
[commit ref fix]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 8c7d6c47a7)
Restarting service causes file-systems to be unmounted without being
mounted back. When this service was obsoleted it should have been
implemented in a way that all actions are ignored. Up to this commit
default handler was called when restart was requested. This default
handler just simply calls stop and start. That means that stop called
unmount but start just printed that this service is obsoleted.
This instead implements restart that just prints same message like start
does. It just calls start in reality. This makes restart unavailable for
call.
Signed-off-by: Karel Kočí <karel.koci@nic.cz>
(cherry picked from commit 3ead9e7b74)
When the non-generic espressobin targets were added these board checks
were not updated. One side effect of this is that config is not saved
during an upgrade of -emmc,-v7, or -v7-emmc devices.
Signed-off-by: Brett Mastbergen <bmastbergen@untangle.com>
(cherry picked from commit 95c1dc5e86)
When targets for multiple ESPRESSObin devices were added, not all
files were updated which means any ESPRESSObin version beside generic
won't have proper networking, sysupgrade and uboot-env. This patch
fixes the issue.
* fixup network detection
* fixup uboot-env
* fixup platform.sh for sysupgrade
Signed-off-by: Vladimir Vid <vladimir.vid@sartura.hr>
(cherry picked from commit bc47285cb3)
Not all versions of ESPRESSObin require SD card, but can
be booted from the internal emmc flash (mmc dev 1) instead.
Add a simple check in the bootscript to see which mmc device
is detected and boot from it using mmcdev variable.
Tested-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
Signed-off-by: Vladimir Vid <vladimir.vid@sartura.hr>
(cherry picked from commit 008b41b9b0)
Just stumbled across this LEDE legacy, without finding any real reason
to keep it. There is a single LEDE_DEVICE_MANUFACTURER_URL dependency
in the luci feed repo which needs to be syncronized.
Signed-off-by: Bjørn Mork <bjorn@mork.no>
[re-added missing commit message]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 8a34a54b6a)
An aquantia phy patch which dropped pause frame support was
missing when integrated LSDK-19.03 kernel patches into OpenWrt.
Without this patch, LS1043A 10G port would not work. So apply it.
Patch link
https://source.codeaurora.org/external/qoriq/qoriq-components/
linux/commit/?h=linux-4.14&id=66346b115818365cfaf99d292871b19f0a1d2850
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
"There are often transient line events when
the system is powered up and initialized and it
is often necessary for the gpio_chip driver to
clear any interrupt flags in hardware before
setting up the gpio chip, especially the
irqchip portions of it."
<http://lists.infradead.org/pipermail/openwrt-devel/2019-June/017630.html>
This patch adds a fix for the APM821XX's interrupt
controller to clear any bogus pending toggled
interrupts that happens on various APM821XX boards
on boot.
The patch also changes the debouce-interval from the
default 5ms debounce interval to 60ms all around.
The default setting caused on occasions that the button
state became stuck in a pressed state, even though the
button was released.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
It seems that newer builds of OpenWRT have a gzip kernel
larger than 2MB~, which for some reason fails to boot on this board.
However, we have 8MB of kernel space and currently the uncompressed
kernel is 6.5MB~, so we have some space to grow until a better
solution is worked out.
Before:
## Booting kernel from Legacy Image at ee840000 ...
Image Name: Linux-4.19.53
Created: 2019-06-22 11:17:48 UTC
Image Type: PowerPC Linux Kernel Image (gzip compressed)
Data Size: 2315724 Bytes = 2.2 MB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
## Loading init Ramdisk from Legacy Image at 02000000 ...
Image Name: OpenWrt fake ramdisk
Created: 2019-06-22 11:17:48 UTC
Image Type: PowerPC Linux RAMDisk Image (uncompressed)
Data Size: 0 Bytes = 0 kB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
## Flattened Device Tree blob at ec000000
Booting using the fdt blob at 0xec000000
Uncompressing Kernel Image ... Error: Bad gzipped data
GUNZIP: uncompress, out-of-mem or overwrite error -
must RESET board to recover
Loading Ramdisk to 10000000, end 10000000 ... OK
Loading Device Tree to 00ffa000, end 00fffc78 ... OK
ft_fixup_l2cache: FDT_ERR_NOTFOUND
After:
## Booting kernel from Legacy Image at ee840000 ...
Image Name: POWERPC OpenWrt Linux-4.19.53
Created: 2019-06-22 11:17:48 UTC
Image Type: PowerPC Linux Kernel Image (uncompressed)
Data Size: 6724584 Bytes = 6.4 MB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
## Loading init Ramdisk from Legacy Image at 02000000 ...
Image Name: OpenWrt fake ramdisk
Created: 2019-06-22 11:17:48 UTC
Image Type: PowerPC Linux RAMDisk Image (uncompressed)
Data Size: 0 Bytes = 0 kB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
## Flattened Device Tree blob at ec000000
Booting using the fdt blob at 0xec000000
Loading Kernel Image ... OK
OK
Loading Ramdisk to 10000000, end 10000000 ... OK
Loading Device Tree to 00ffa000, end 00fffc78 ... OK
Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com> [75 cpl limit]
When OEM volumes are present in the [alt_]firmware partition,
sysupgrade will write a new kernel, but will fail to write
the root file system. The next boot will hang indefinitely
Waiting for root device /dev/ubiblock0_0...
Modified ipq40xx/base-files/lib/upgrade/linksys.sh
to remove both `squashfs` and `ubifs` if found
on the target firmware partition's UBI device.
Run-tested-on: Linksys EA8300
Signed-off-by: Jeff Kletsky <git-commits@allycomm.com>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
[applied some shellcheck suggestions as well]
Follow upstream changes - header file changes only
no functional or executable changes, hence no package bump
required
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit a8f0c02f80)
Follow upstream changes - header file changes
no executable difference at all
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 5273fb6b21)
This patch adds a missing type property which prevented
the creation of oneshot and timer led triggers when they
are specified in the /etc/board.d/01_leds files.
i.e.:
ucidef_set_led_timer "system" "system" "zhuotk:green:system" "1000" "1000"
Fixes: b06a286a48 ("base-files: cleanup led functions in uci-defaults.sh")
Signed-off-by: Robinson Wu <wurobinson@qq.com>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Move all MikroTik devices to new function to increase script execution
speed.
Machine name in new version of MikroTik RouterBOARD devices add "RB"
before model name:
Old machine name: MikroTik RouterBOARD 951Ui-2nD
New: MikroTik RouterBOARD RB951Ui-2nD
So this patch should fix it for all currently supported MikroTik boards.
Tested-by: Cezary Jackiewicz <cezary@eko.one.pl> [wAP R-2nD]
Signed-off-by: Henryk Heisig <hyniu@o2.pl>
[rebased,commit message facelift,script fixes]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
[spotted missing 922UAGS-5HPacD]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Refreshed all patches.
Altered patches:
- 816-pcie-support-layerscape.patch
This patch also restores the initial implementation
of the ath79 perfcount IRQ issue. (78ee6b1a40)
It was wrongfully backported upstream initially and got reverted now.
Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
err_free_stats has been deprecated. Replace with err_netdev.
Compile-tested on: mvebu
Runtime-tested on: mvebu
Fixes: 4089df4f4b ("kernel: bump 4.14 to 4.14.125 (FS#2305 FS#2297)")
Signed-off-by: George Amanakis <gamanakis@gmail.com>
[altered hashes]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 1e3800df18)
It's needed for applying some hardware quirks. This fixes:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/dmi.c:60:20: error: 'DMI_PRODUCT_SKU' undeclared here (not in a function); did you mean 'DMI_PRODUCT_UUID'?
DMI_EXACT_MATCH(DMI_PRODUCT_SKU, "T8"),
Fixes: c52054e568 ("mac80211: brcm: backport remaining brcmfmac 5.2 patches")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 4d11c4c378)
This improves FullMAC firmware compatibility, adds logging in case of
firmware crash and *may* fix "Invalid packet id" errors.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 8888cb725d)
FCC allows maximum antenna gain of 6 dBi. 15.247(b)(4):
> (4) The conducted output power limit
> specified in paragraph (b) of this section
> is based on the use of antennas
> with directional gains that do not exceed
> 6 dBi. Except as shown in paragraph
> (c) of this section, if transmitting
> antennas of directional gain greater
> than 6 dBi are used, the conducted
> output power from the intentional radiator
> shall be reduced below the stated
> values in paragraphs (b)(1), (b)(2),
> and (b)(3) of this section, as appropriate,
> by the amount in dB that the
> directional gain of the antenna exceeds
> 6 dBi.
https://www.gpo.gov/fdsys/pkg/CFR-2013-title47-vol1/pdf/CFR-2013-title47-vol1-sec15-247.pdf
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Most of the txpower for the ath10k firmware is stored as twicepower (0.5 dB
steps). This isn't the case for max_antenna_gain - which is still expected
by the firmware as dB.
The firmware is converting it from dB to the internal (twicepower)
representation when it calculates the limits of a channel. This can be seen
in tpc_stats when configuring "12" as max_antenna_gain. Instead of the
expected 12 (6 dB), the tpc_stats shows 24 (12 dB).
Tested on QCA9888 and IPQ4019 with firmware 10.4-3.5.3-00057.
Signed-off-by: Sven Eckelmann <sven@narfation.org>
FCC allows maximum antenna gain of 6 dBi. 15.247(b)(4):
> (4) The conducted output power limit
> specified in paragraph (b) of this section
> is based on the use of antennas
> with directional gains that do not exceed
> 6 dBi. Except as shown in paragraph
> (c) of this section, if transmitting
> antennas of directional gain greater
> than 6 dBi are used, the conducted
> output power from the intentional radiator
> shall be reduced below the stated
> values in paragraphs (b)(1), (b)(2),
> and (b)(3) of this section, as appropriate,
> by the amount in dB that the
> directional gain of the antenna exceeds
> 6 dBi.
https://www.gpo.gov/fdsys/pkg/CFR-2013-title47-vol1/pdf/CFR-2013-title47-vol1-sec15-247.pdf
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Most of the txpower for the ath10k firmware is stored as twicepower (0.5 dB
steps). This isn't the case for max_antenna_gain - which is still expected
by the firmware as dB.
The firmware is converting it from dB to the internal (twicepower)
representation when it calculates the limits of a channel. This can be seen
in tpc_stats when configuring "12" as max_antenna_gain. Instead of the
expected 12 (6 dB), the tpc_stats shows 24 (12 dB).
Tested on QCA9888 and IPQ4019 with firmware 10.4-3.5.3-00057.
Signed-off-by: Sven Eckelmann <sven@narfation.org>
FCC allows maximum antenna gain of 6 dBi. 15.247(b)(4):
> (4) The conducted output power limit
> specified in paragraph (b) of this section
> is based on the use of antennas
> with directional gains that do not exceed
> 6 dBi. Except as shown in paragraph
> (c) of this section, if transmitting
> antennas of directional gain greater
> than 6 dBi are used, the conducted
> output power from the intentional radiator
> shall be reduced below the stated
> values in paragraphs (b)(1), (b)(2),
> and (b)(3) of this section, as appropriate,
> by the amount in dB that the
> directional gain of the antenna exceeds
> 6 dBi.
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Memory is allocated with devm_kzalloc() on every page program
and leaks until device is closed (which never happens).
Convert to kzalloc() and handle error paths manually.
Signed-off-by: Mantas Pucka <mantas@8devices.com>
Refreshed all patches.
This bump contains upstream commits which seem to avoid (not properly fix)
the errors as seen in FS#2305 and FS#2297
Altered patches:
- 403-net-mvneta-convert-to-phylink.patch
- 410-sfp-hack-allow-marvell-10G-phy-support-to-use-SFP.patch
Compile-tested on: ar71xx, cns3xxx, imx6, mvebu, x86_64
Runtime-tested on: ar71xx, cns3xxx, imx6, x86_64
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* Remove testing patches for kernel version 4.19
* remove targets ar7, ixp4xx, orion
Those targets are still on kernel 4.9, patches for 4.14 were not ready
in time. They may be readded once people prepare and test patches for
kernel 4.14.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
@@ -8390,6 +8391,7 @@ int ath10k_mac_register(struct ath10k *a
@@ -8394,6 +8395,7 @@ int ath10k_mac_register(struct ath10k *a
ar->hw->wiphy->bands[NL80211_BAND_5GHZ] = band;
}
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
