DENG Qingfang [Thu, 30 Apr 2020 11:41:39 +0000 (19:41 +0800)]
mvebu: use eth0 as DSA CPU port for Linksys WRT
eth0 has HW MAC address while eth2 does not.
Use eth0 instead so we don't have to set LAN MAC manually.
Disable unused eth2, until multi CPU port is supported.
Signed-off-by: DENG Qingfang <dengqf6@mail2.sysu.edu.cn>
This is useful when booting OpenWrt from ramdisks in order to have both
images partitions defined.
Furthermore, instead of always using img2 for the inactive image, let's use
img1 or img2 accordingly.
Tobias Schramm [Wed, 3 Jun 2020 13:10:40 +0000 (15:10 +0200)]
ramips: mt7621: use higher SPI clock speed on Mikrotik rbm11g and rbm33g
Previously the dts were using a value determined by empirical testing,
because of a spi driver/clock bug. The bug was fixed quite some time
ago. 33 MHz is the default clock frequency used by RouterBOOT and thus
safe.
Tim Harvey [Thu, 28 May 2020 14:58:49 +0000 (07:58 -0700)]
octeontx: fix mcp251x can controller
Update the can-mcp251x-convert-to-half-duplex-SPI patch to fix reception
Some SPI host controllers such as the Cavium Thunder TX do not support
full-duplex SPI. Using half-duplex transfers allows the driver to work
with those host controllers.
This patch fixes the fact that mcp251x_hw_rx_frame was still relying on
a full-duplex transfer where bits were being shifted on MOSI at the same time
as MISO. After splitting the transaction into a spi_write_then_read() care
must be taken to ignore the first byte.
Tim Harvey [Thu, 28 May 2020 13:56:17 +0000 (06:56 -0700)]
imx6: backport v5.8 imx6qdl-gw dt patches
- add fxos8700 support to GW52xx/GW53xx/GW54xx
- add USB_OTG support to GW552x
- add LSM9DS1 IMU support to GW560x
- add LSM9DS1 IMU support to GW5904
- add CC1352 UART to GW5910
- add BCM4330 support to GW5910
- fix wlan regulator for GW5910
wolfssl: use -fomit-frame-pointer to fix asm error
32-bit x86 fail to compile fast-math feature when compiled with frame
pointer, which uses a register used in a couple of inline asm functions.
Previous versions of wolfssl had this by default. Keeping an extra
register available may increase performance, so it's being restored for
all architectures.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Since this change I am getting the following error message on boot.
> kern.err kernel: [ 1.009416] gpio_it87: no device
I do not have this device on my system. To prevent the nonsensical
message and the loading of the module I have added this as a package, so
that it can be installed later or during image building.
Signed-off-by: Florian Eckert <fe@dev.tdt.de> Reviewed-by: Philip Prindeville <philipp@redfish-solutions.com>
base-files: prevent issues w/ overlay on powerloss after sysupgrade
Due to filesystem write caching the old configuration data could stay
out of flash for a long time during a first boot after the sysupgrade.
Power loss during this period could damage the overlay data and even
make device inaccessable via the network.
Fix this by syncing data to a flash as soon as the previous
configuration will be unpacked after the sysupgrade. Also sync the FS
state after the sysupgrade.tgz archive removing to prevent duplicative
extraction of a previous configuration.
mvebu: espressobin: fix SATA and USB 3.0 ports detection
This commit removes changes from upstream commits: 8e18c8e58da6 arm64: dts: marvell: armada-3720-espressobin: declare SATA
PHY property bd3d25b07342 arm64: dts: marvell: armada-37xx: link USB hosts with their
PHYs
For most boards which have factory bootloader this caused that devices
connected to USB 3.0 and SATA port were not detected. For them to
function users would need to upgrade the bootloader to version with ARM
Trusted Firmware 2.1 or later. Unfortunately there is no official
bootloader image with updated ATF component, therefore drop these
properties from nodes. This change was also tested briefly with
bootloader with updated ATF and the ports functioned properly.
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
Thomas Albers [Wed, 3 Jun 2020 11:42:47 +0000 (13:42 +0200)]
base-files: fix LED IDE trigger
This changes the ide-disk LED trigger to the generic disk-activity as
ide-disk trigger was removed in upstream commit eb25cb9956cc ("leds:
convert IDE trigger to common disk trigger").
Signed-off-by: Thomas Albers <thomas.gameiro@googlemail.com>
[split into separate commit, commit description facelift] Signed-off-by: Petr Štetiar <ynezz@true.cz>
Kernel config option LEDS_TRIGGER_IDE_DISK was renamed in kernel 4.8 to
CONFIG_LEDS_TRIGGER_DISK in upstream commit eb25cb9956cc ("leds: convert
IDE trigger to common disk trigger").
Removing it as it should be added only on targets which has usage for
this trigger.
Signed-off-by: Thomas Albers <thomas.gameiro@googlemail.com>
[commit description facelift] Signed-off-by: Petr Štetiar <ynezz@true.cz>
In FS#2738 we can see that patch first introduced in e8ebcff ("ramips: add a explicit reset to dwc2")
breaks USB functionality since 18.06. Thus, this patch should be removed.
Rosen Penev [Mon, 30 Mar 2020 01:13:56 +0000 (18:13 -0700)]
prereq-build: test for perl's Data::Dumper
Required for installation of autoconf:
make[5]: Entering directory `/openwrt/build_dir/host/autoconf-2.69'
Making all in bin
make[6]: Entering directory `/openwrt/build_dir/host/autoconf-2.69/bin'
autom4te_perllibdir='..'/lib AUTOM4TE_CFG='../lib/autom4te.cfg'
../bin/autom4te -B '..'/lib -B '..'/lib --language M4sh --cache
'' --melt ./autoconf.as -o autoconf.in
Can't locate Data/Dumper.pm in @INC (@INC contains: ../lib
/usr/local/lib64/perl5 /usr/local/share/perl5 /usr/lib64/perl5/vendor_perl
/usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5 .) at
../lib/Autom4te/C4che.pm line 33.
BEGIN failed--compilation aborted at ../lib/Autom4te/C4che.pm line 33.
Compilation failed in require at ../bin/autom4te line 40.
BEGIN failed--compilation aborted at ../bin/autom4te line 40.
make[6]: *** [autoconf.in] Error 2
apm821xx: set DEVICE_TYPE to "nas" for sata subtarget
Since DEVICE_TYPE cannot be set per device, just set DEVICE_TYPE
to "nas" for the entire subtarget, which only contains this single
device.
Note that while this looks like a cosmetic change in combination
with the previous patches, this particular patch actually changes
the packages for the device.
Suggested-by: Christian Lamparter <chunkeey@gmail.com> Cc: Christian Lamparter <chunkeey@gmail.com> Cc: Sungbo Eo <mans0n@gorani.run> Cc: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
While the effective "default" based on frequent use is "router", the
DEVICE_TYPE variable actually provides a "basic" configuration without
selecting any additional packages.
This is currently set up with the identifier "bootloader", which seems
to be not used at all. However, the only targets not using "router" or
"nas" are actually archs38 and arc770, which use their own value
"developerboard" for DEVICE_TYPE which seems to have been invented when
these targets where added. The latter is not implemented in target.mk,
though, and will fall back to the "basic" set of packages then.
So, to clean this up and make it more readable, let's just define a
DEVICE_TYPE "basic" and use it for the aforementioned cases.
Cc: Christian Lamparter <chunkeey@gmail.com> Cc: Sungbo Eo <mans0n@gorani.run> Cc: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
treewide: drop DEVICE_TYPE when used as device variable
DEVICE_TYPE is a target/subtarget variable, and it does not have
any effect when set in a device definition. It can only be set
in a target's or subtarget's Makefile.
Consequently, having it set anyway is misleading, so this drops
all cases.
This effectively reverts the following commits: 7a1497fd601d ("apm821xx: MBL: set DEVICE_TYPE to NAS") 5b4765c93a1b ("gemini: Classify Raidsonic NAS IB-4220-B as a NAS") cdc6de460bb4 ("gemini: D-Link DNS-313 is a NAS")
For the following commit, the variable was set when adding device
support: 27b2f0fc0fc5 ("kirkwood: add support for Iomega Storcenter ix2-200")
Cc: Christian Lamparter <chunkeey@gmail.com> Cc: Sungbo Eo <mans0n@gorani.run> Cc: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Some (older) CFEs are loaded at 0x80401000 and ramdisks are loaded at
0x80010000, which means that ramdisk size limit is 0x3F1000 (almost 4M).
Therefore, current ramdisks (~4MB) are overwritting CFE in these devices,
which results in a crash.
This commit changes the address where ramdisks are loaded to 0x80a00000,
which is the same address where kernel is loaded when booting from the flash.
Therefore, lzma-loader will now be loaded at 0x80a00000, but it will still
decompress the kernel at 0x80010000.
Tested with huawei,hg556a-b, which has its CFE loaded at 0x80401000.
Apparently, Sercomm allows loading a BCM WFI image via CFE, but this image
destroys "serial" and "protect" nand partitions, which is wrong.
It will also set both bootflags to the same value, which causes booting
issues with cferam (cferom will alternatively boot from cferam1 or cferam2
each time the device is rebooted).
Now that OEM Sercomm images are supported it's time to remove this hacky
cfe.bin image support.
BCM6368 and newer devices are compatible with any lzma compression parameters.
Add a new legacy device definition and use it on BCM6358 and older devices.
When firmware is flashed, cferam.000 extension is renamed to the next number.
When booting, CFE scans the NAND and picks the partition with the highest
cferam extension and ignores the other one.
Support Sercomm firmware partition split.
WFI partition must be defined after bootflag partitions in order for the
parser to properly find bootflag1 and bootflag2 partitions.
kernel: mtdsplit: bcm_wfi: support Sercomm bootflags
Sercomm uses 2 bootflag partitions and boots the firmware with the highest
bootflag. Support splitting the firmware partition while keeping support for
unsplitted layout.
Sercomm firmwares are encrypted with AES 256 CBC.
The key is generated with a custom algorithm from the firmware tag:
char key[32];
char version[32];
char iv[32];
char random[32];
char size[32];
Key must be generated with Sercomm's algorithm. However, the rest of the
header can be empty. IV and random are set to 0 on purpose.
Hans Dedecker [Sun, 31 May 2020 18:28:06 +0000 (20:28 +0200)]
ppp: update to version 2.4.8.git-2020-05-25
ddd57c2 pppd: Add lcp-echo-adaptive option c319558 pppd: Handle SIGINT and SIGTERM during interrupted syscalls (#148) 0bc11fb Added missing options to manual pages. (#149) b1fcf16 Merge branch 'monotonic-time' of https://github.com/themiron/ppp c78e312 pppd: linux: use monotonic time if possible
Remove patch 121-debian_adaptive_lcp_echo as patch is upstream accepted
Remove patch 206-compensate_time_change.patch as timewrap issues are
solved by a patch making use of monotonic time
David Bauer [Sat, 30 May 2020 14:24:03 +0000 (16:24 +0200)]
ar71xx: correct button type for TL-MR3020 mode slider
The TP-Link TL-MR3020 has a three-state mode slider which was previously
integrated as a button (EV_KEY). This led to spurious activations of
failsafe mode.
Set the type for the button to switch (EV_SW), to avoid unintended
activations of failsafe mode.
Related: commit 27f3f493de06 ("gpio-button-hotplug: unify polled and
interrupt code")
Andreas Wiese [Wed, 27 May 2020 13:31:30 +0000 (15:31 +0200)]
ath79: add support for TP-Link RE450 v3
TP-Link RE450 v3 is a dual band router/range-extender based on
Qualcomm/Atheros QCA9563 + QCA9880.
This device is nearly identical to RE450 v2 besides a modified flash
layout (hence I think force-flashing a RE450v2 image will lead to at
least loss of MAC address).
Specification:
- 775 MHz CPU
- 64 MB of RAM (DDR2)
- 8 MB of FLASH (SPI NOR)
- 3T3R 2.4 GHz
- 3T3R 5 GHz
- 1x 10/100/1000 Mbps Ethernet (AR8033 PHY)
- 7x LED, 4x button-
- possible UART header on PCB¹
Flash instruction:
Apply factory image in OEM firmware web-gui.
¹ Didn't check to connect as I didn't even manage to connect on
RE450v2 (AFAIU it requires disconnecting some resistors, which I was
too much of a coward to do). But given the similarities to v2 I
think it's the same or very similar procedure (and most likely also
the only way to debrick).
Signed-off-by: Andreas Wiese <aw-openwrt@meterriblecrew.net>
[remove dts-v1 and compatible in DTSI] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Lech Perczak [Fri, 29 May 2020 18:50:05 +0000 (20:50 +0200)]
ath79: add support for TP-Link TL-WR802N V1 and V2
Specification:
- SoC: Qualcomm Atheros QCA9533 (560 MHz, MIPS 24Kc)
- RAM: 32 MiB
- Storage: 4 MiB of Flash on board
- Wireless: Built into QCA9533 (Honey Bee), PHY modes b/g/n
- Ethernet: 1x100M (port0)
Installation through OEM Web Interface:
- Connect to TL-WR802N by Ethernet or Wi-Fi
- Go to web interface:
[V1] http://192.168.0.1
[V2] http://192.168.0.254
Default user is "admin" & password is "admin".
On V2, there is no DHCP server running by default, so remember to set
IP manually.
- Go to "System Tools -> Firmware Upgrade"
- Browse for firmware:
[V1] "*.factory.bin"
[V2] "*.factory-us.bin" or "*.factory-eu.bin" for eu model
Web interface may complain if filename is too long. In such case,
rename .bin to something shorter.
- Click upgrade
Installation through tftp:
Note: T_OUT, T_IN and GND on the board must be connected to USB TTL
Serial Configuration 115200 8n1
- Boot the TL-WR802N
- When "Autobooting in 1 seconds" appears type "tpl" followed by enter
- Connect to the board Ethernet port
(IPADDR: 192.168.1.1, ServerIP: 192.168.1.10)
- tftpboot 0x80000000 <Firmware Image Name>
- Record the result of "printenv bootcmd"
- Enter "erase <Result of 'printenv bootcmd'> +0x3c0000"
(e.g erase 0x9f020000 +0x3c0000)
- Enter "cp.b 0x80000000 <Result of 'printenv bootcmd'> 0x3c0000"
(e.g cp.b 0x80000000 0x9f020000 0x3c0000)
- Enter "bootm <Result of 'printenv bootcmd'>"
(e.g bootm 0x9f020000)
Notes:
When porting from ar71xx target to ath79, I found out that on V2,
reset button is on GPIO12 and active low, instead of GPIO11 and
active high. By cross-flashing V1 firmware to V2, I confirmed
the same is true for V1.
Also according to manual of V1, this one also has green
LED instead of blue - both of those issues were fixed accordingly.
The MAC address assignment has been checked with OEM firmware.
Installation manual based on ar71xx support by Thomas Roberts
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
[slightly adjust commit message, add MAC address comment] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Lech Perczak [Fri, 29 May 2020 19:56:18 +0000 (21:56 +0200)]
ar71xx: fix reset key for TP-Link TL-WR802N V1/V2
During porting support for this router to ath79 target
it was discovered that GPIO mapping was incorrect (GPIO11 active high).
Correct mapping for both V1 and V2 is GPIO12 active low.
Default configuration from GPL source for V2 explicitly states this, and
this was confirmed experimentally on ath79 by looking on
/sys/kernel/debug/gpio. Correctness of this was also validated for V1 by
cross-flashing vendor firmware for V1 on V2 hardware, in which reset
button also worked.
zstd with its default settings (compression level -3) compresses better
than bzip2 -9 (which is the default setting), and is an order of magnitude
faster.
I made the following measurements for the most common compression tools
(all standard Debian Buster versions, default flags unless noted
otherwise), using the debug information of a large x86-64 kernel with
ALL_KMODS:
* kernel-debug.tar: 376M
* kernel-debug.tar.gz: 101M, compressed in ~12s
* kernel-debug.tar.bz2: 91M, compressed in ~15s
* kernel-debug.tar.xz: 57M, compressed in ~101s
* kernel-debug.tar.zst: 86M, compressed in ~1s
With zstd, there is still some room for improvement by increasing the
compression, but the slight increase in compression ratio
(22.83% -> 19.46%) does not justify the significant increase in
compression time (about 5 times on my machine) in my opinion.
Note that multithreaded compression (-T argument) does not affect
reproducibility with zstd.
- add Ed25519 support (backport):
* DROPBEAR_ED25519 option for ssh-ed25519,
* disabled by default
- add Chacha20-Poly1305 support (backport):
* DROPBEAR_CHACHA20POLY1305 for chacha20-poly1305@openssh.com,
* enabled by default
- update feature costs in binary size
Daniel Golle [Sat, 30 May 2020 08:36:17 +0000 (09:36 +0100)]
mac80211: rt2x00: backport patch enabling MFP
From: Rui Salvaterra <rsalvaterra@gmail.com>
Date: Mon, 25 May 2020 14:49:07 +0100
Subject: [PATCH] rt2800: enable MFP support unconditionally
This gives us WPA3 support out of the box without having to manually disable
hardware crypto. The driver will fall back to software crypto if the connection
requires management frame protection.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Fri, 29 May 2020 22:58:20 +0000 (23:58 +0100)]
ugps: nmea: make sure date is valid
GPS time without date was previously used to set system date:
Tue Oct 10 11:48:21 2000 user.info kernel: [ 108.786639] ugps: system time differs from GPS time by more than 5 seconds. Using 2000-10-10T10:48:21 UTC as the new time
Tue Oct 10 11:49:27 2000 user.info kernel: [ 174.794699] ugps: system time differs from GPS time by more than 5 seconds. Using 2020-05-26T10:49:27 UTC as the new time
Fix this by ignoring incomplete dates and wait for complete time
information before adjusting system date/time.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
ar71xx: fix splitting firmware partition for TL-WR902AC v1
The -O option for the tplink-v1-header was missing for the TP-Link
TL-WR902AC v1, while safeloader and MTDPARTS where set up with a
single firmware partition.
This led to bootloops after using sysupgrade.
Fixes: FS#3118 Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Jo-Philipp Wich [Thu, 28 May 2020 11:03:02 +0000 (13:03 +0200)]
broadcom-wl: don't inherit lock descriptor in nas process
Add a local hack to prevent the Broadcom WPA authenticator process from
inheriting the lock descriptor 1000 used to prevent concurrent executions
of the init script.
Without this fix, repeated invocations of /etc/init.d/network, e.g. for
obtaining the enabled state, would hang forever.
bcm27xx-userland: update to latest version with 64 bit support
Support for 64 bits has been remove on latest master of raspberry/firmware.
Update to latest commit with 64 bit support since we don't support
installing 32 bit packages on 64 bit targets.
Jakov Petrina [Mon, 25 May 2020 12:14:40 +0000 (14:14 +0200)]
linux: mvebu: backport mvneta XDP support fixes
This patch backports additional fixes for XDP support in the mvneta driver. These
changes are found upstream as commits:
b37fa92e20ef2 net: mvneta: fix build skb for bm capable devices f383b2950070c net: mvneta: rely on page_pool_recycle_direct in mvneta_run_xdp 79572c98c554d mvneta driver disallow XDP program on hardware buffer management 44efc78d0e464 net: mvneta: fix XDP support if sw bm is used as fallback
Thibaut VARÈNE [Thu, 21 May 2020 15:40:54 +0000 (17:40 +0200)]
packages/utils: fbtest fix Makefile
The clean target tries to remove what looks like a bogus 'rbcfg',
probably carried over copy-pasta. Remove the name of the generated
executable ('fbtest') instead.
Thibaut VARÈNE [Sat, 16 May 2020 20:18:15 +0000 (22:18 +0200)]
package/base-files: caldata: work around dd's limitation
tl;dr: dd will silently truncate the output if reading from special
files (e.g. sysfs attributes) with a too large bs parameter.
This problem was exposed on some RouterBOARD ipq40xx devices which use a
caldata payload which is larger than PAGE_SIZE, contrary to all other
currently supported RouterBOARD devices: the caldata would fail to
properly load with the current scripts.
Background: dd doesn't seem to correctly handle read() results that
return less than requested data. sysfs attributes have a kernel exchange
buffer which is at most PAGE_SIZE big, so only 1 page can be read() at a
time. In this case, if bs is larger than PAGE_SIZE, dd will silently
truncate blocks to PAGE_SIZE. With the current scripts using bs=<size>
count=1, the data is truncated to PAGE_SIZE as soon as the requested
<size> exceeds this value.
This commit works around this problem by using `cat` in the caldata
routines that can read from a file (routines that read from mtd devices
are untouched). cat correctly handles partial read requests. The output
is then piped to dd with the same parameters as before, to ensure that
the resulting file remains exactly the same.
This is a simple workaround, the downside is that it uses a pipe and one
more executable, and therefore has a larger memory footprint and is
slower. This is deemed acceptable considering these routines are only
used at boot time.
Tested-by: Robert Marko <robimarko@gmail.com> Signed-off-by: Thibaut VARÈNE <hacks@slashdirt.org>
projects / thirdparty / openwrt.git / log
