openwrt/staging/lynxis.git
3 years agoqosify: add missing dependency
Felix Fietkau [Mon, 8 Nov 2021 20:28:27 +0000 (21:28 +0100)]
qosify: add missing dependency

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agobcm4908: fix b53 regression regarding RGMII reg(s)
Rafał Miłecki [Mon, 8 Nov 2021 16:23:09 +0000 (17:23 +0100)]
bcm4908: fix b53 regression regarding RGMII reg(s)

This fixes:
[   35.308550] WARNING: CPU: 0 PID: 230 at bcm_sf2_reg_rgmii_cntrl.isra.0+0x84/0x90

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agobcm53xx: drop patch for downstream swconfig-based b53
Rafał Miłecki [Mon, 8 Nov 2021 14:58:42 +0000 (15:58 +0100)]
bcm53xx: drop patch for downstream swconfig-based b53

It isn't used anymore so there is no need to hack CPU port. Upstream
(DSA-based) b53 also supports all switch ports just fine.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agobcm53xx: drop downstream patch for Buffalo WZR-1750DHP button
Rafał Miłecki [Thu, 16 Sep 2021 06:39:50 +0000 (08:39 +0200)]
bcm53xx: drop downstream patch for Buffalo WZR-1750DHP button

This patch never received a proper description and was never sent
upstream as supposed. It was meant to be handled in 2015-2018 so it
should be safe to assume noone really understands it or care.

Cc: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agolayerscape: mkits-multiple-config: fail on invalid arguments
Rafał Miłecki [Mon, 25 Oct 2021 08:17:15 +0000 (10:17 +0200)]
layerscape: mkits-multiple-config: fail on invalid arguments

Fail if arguments couldn't be parsed and print unrecognized part. It's
important when running it from script with dynamic values. Missing value
could result in skipping argument and silent failures (unexpected its).

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agobcm4908: start work on images for devices using U-Boot
Rafał Miłecki [Wed, 3 Nov 2021 17:12:05 +0000 (18:12 +0100)]
bcm4908: start work on images for devices using U-Boot

New BCM4908 family based routers will use U-Boot bootloader. That will
require using a totally different firmware format. Kernel has to be put
in a FIT image.

OpenWrt has some helpers for generating .its files but they don't fit
BCM4908 requirements and there is no simple way of extending any of
them. The best solution seems to be storing an .its template.

BCM4908 bootfs may:
1. contain extra binaries (other than kernel & DTB)
2. include multiple DTB files
3. store device specific U-Boot configurations with custom properties

Such setups are too complex to generate using shell script. Raw .its
file on the other hand seems quire clean & reasonable.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agobcm47xx: switch to kernel 5.10
Rafał Miłecki [Mon, 8 Nov 2021 14:56:39 +0000 (15:56 +0100)]
bcm47xx: switch to kernel 5.10

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agobcm47xx: fix bgmac regression present in 5.10 kernel
Rafał Miłecki [Mon, 8 Nov 2021 14:55:40 +0000 (15:55 +0100)]
bcm47xx: fix bgmac regression present in 5.10 kernel

This fixes:
[    2.548098] bgmac_bcma bcma0:1: Failed to register fixed PHY device
[    2.554584] bgmac_bcma bcma0:1: Cannot connect to phy
and downstream (swconfig-based) b53 driver failing to load.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agobcm47xx: relocate LZMA loader
Rafał Miłecki [Sun, 7 Nov 2021 13:45:14 +0000 (14:45 +0100)]
bcm47xx: relocate LZMA loader

Kernel 5.10 grew bigger than 5.4 so we need to bump BZ_TEXT_START to
allow lzma loader hanel its size.

At the same time BZ_STACK_START needs to be increased to avoid
overwriting the stack.

For a reference see:
d5cf4a5aa4a3 ("brcm47xx: relocate loader to higher address")
2909a4b78e2b ("brcm47xx: relocate the stack in loader")

Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agorpcd: reload rpcd on installation of rpcd-mod-*
Daniel Golle [Fri, 29 Oct 2021 11:14:16 +0000 (12:14 +0100)]
rpcd: reload rpcd on installation of rpcd-mod-*

When installing additional rpcd modules, a restart of rpcd is required.
This often confuses users as even after installing rpcd-mod-rpcsys the
relevant ubus objects are still missing until rpcd has been reloaded
(or the system has been rebooted, obviously).
Let rpcd-mod-* reload rpcd as post-install action.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
3 years agoqosify: update to the latest version
Felix Fietkau [Mon, 8 Nov 2021 14:04:57 +0000 (15:04 +0100)]
qosify: update to the latest version

2ca7352543da map: make a helper function for freeing entries
411432ec853b map: add support for adding dns regex patterns
14803cb559d8 ubus: remove unused enum
a0740172eda6 ubus: add api for providing dns lookup results for dns regex rules
406fbf478e87 ubus: add support for dynamically adding dns based rules
5fc91183d60a README: mention dns regex entries
3ed8c3eb1a3b README: document mapping file syntax
91ce2e77d302 map: introduce low effort codepoint from RFC8622
5ff14acca0e7 interface: enable NAT on interfaces by default
e70f70e496d7 README: fix typo
f25ded617478 README: fix another typo
675238bc2ce5 loader: always reinitialize programs
010eea0d98c3 map: improve timeout handling of IP entries
7ef54a7f04a0 map: add DF codepoint
6f7fbe698555 map: increase active timeout to 300
60e06a579a13 qosify-bpf: inline check_flow() to ensure that it is jited
f5ae89e8d869 ubus: subscribe to dnsmasq.dns for dns lookup results

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agodnsmasq: add support for monitoring and modifying dns lookup results via ubus
Felix Fietkau [Mon, 8 Nov 2021 13:48:36 +0000 (14:48 +0100)]
dnsmasq: add support for monitoring and modifying dns lookup results via ubus

The monitoring functionality will be used for dns rule support in qosify

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agokernel: bump 5.10 to 5.10.78
John Audia [Sat, 6 Nov 2021 16:31:50 +0000 (12:31 -0400)]
kernel: bump 5.10 to 5.10.78

All patches automatically rebased.

Build system: x86_64
Build-tested: bcm2711/RPi4B
Run-tested: bcm2711/RPi4B

Signed-off-by: John Audia <graysky@archlinux.us>
3 years agokernel: bump 5.4 to 5.4.158
John Audia [Sat, 6 Nov 2021 16:34:16 +0000 (12:34 -0400)]
kernel: bump 5.4 to 5.4.158

All patches automatically rebased.

Signed-off-by: John Audia <graysky@archlinux.us>
3 years agogdb: Make only full gdb depend on libgmp
Hauke Mehrtens [Sun, 7 Nov 2021 16:15:06 +0000 (17:15 +0100)]
gdb: Make only full gdb depend on libgmp

libgmp is only needed for the full gdb and not for the gdbserver
application.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agogdb: Add explicit patch to libgmp
Hauke Mehrtens [Sun, 7 Nov 2021 16:14:16 +0000 (17:14 +0100)]
gdb: Add explicit patch to libgmp

Without giving the patch gdb does not compile on Arch Linux.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agopistachio: drop Kernel 5.4 support
Paul Spooren [Wed, 3 Nov 2021 00:41:23 +0000 (14:41 -1000)]
pistachio: drop Kernel 5.4 support

With the switch to Kernel 5.10 the 5.4 files are no longer needed.

Signed-off-by: Paul Spooren <mail@aparcar.org>
3 years agopistachio: switch to Kernel 5.10
Paul Spooren [Wed, 3 Nov 2021 00:40:08 +0000 (14:40 -1000)]
pistachio: switch to Kernel 5.10

This was successfully boot tested on a Creator Ci40.

Signed-off-by: Paul Spooren <mail@aparcar.org>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agokernel: fix KERNEL_KASAN_VMALLOC build option
Stijn Tintel [Sun, 7 Nov 2021 16:57:12 +0000 (18:57 +0200)]
kernel: fix KERNEL_KASAN_VMALLOC build option

It should be config, not CONFIG.

Fixes: cbeab2cd9919 ("kernel: add another missing KASAN symbol")
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
3 years agokernel: add missing keyword to KERNEL_KASAN_VMALLOC
Stijn Tintel [Sun, 7 Nov 2021 16:54:28 +0000 (18:54 +0200)]
kernel: add missing keyword to KERNEL_KASAN_VMALLOC

The help keyword is missing, which breaks menuconfig etc.

Fixes: cbeab2cd9919 ("kernel: add another missing KASAN symbol")
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
3 years agokernel: add another missing KASAN symbol
Stijn Tintel [Sun, 7 Nov 2021 16:33:35 +0000 (18:33 +0200)]
kernel: add another missing KASAN symbol

Enabling KERNEL_KASAN exposes yet another missing symbol. This did not
appear on bcm27xx but is appearing now on x86/64. Add a new kernel build
option for KASAN_VMALLOC to fix build on x86/64 with KERNEL_KASAN
enabled.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agoarchs38: update kernel version to 5.10
Evgeniy Didin [Tue, 2 Nov 2021 12:45:01 +0000 (15:45 +0300)]
archs38: update kernel version to 5.10

Update config, Makefile and image/Makefile.
Directly switch to kernel 5.10.

This patch was tested in nSIM simulator, no errors appeared.

Signed-off-by: Evgeniy Didin <Evgeniy.Didin@synopsys.com>
Cc: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Paul Spooren <mail@aparcar.org>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Cc: John Crispin <john@phrozen.org>
3 years agokernel: Add regmap-i2c dependency to sound-soc-imx-sgtl5000
Hauke Mehrtens [Sat, 6 Nov 2021 18:56:20 +0000 (18:56 +0000)]
kernel: Add regmap-i2c dependency to sound-soc-imx-sgtl5000

This dependency is needed on the imx/cortexa7 subtarget.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agokernel: add missing UBSAN config symbols
Stijn Tintel [Wed, 3 Nov 2021 17:38:50 +0000 (19:38 +0200)]
kernel: add missing UBSAN config symbols

Enabling KERNEL_UBSAN exposes several missing symbols. Add new kernel
build options for UBSAN_BOUNDS and UBSAN_TRAP, disable CONFIG_TEST_UBSAN
in the generic kernel configs and enable CONFIG_UBSAN_MISC in generic
5.10 config. The latter symbol was removed in later kernels, as it was
causing some issues, so just disable it in 5.10 instead of adding a
build option for it.

Fixes build failures with KERNEL_UBSAN enabled.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agokernel: add missing KASAN config symbols
Stijn Tintel [Wed, 3 Nov 2021 18:00:26 +0000 (20:00 +0200)]
kernel: add missing KASAN config symbols

Enabling KERNEL_KASAN exposes several missing symbols. As KASAN_SW_TAGS
is only implemented for arm64 CPUs and requires clang, it doesn't make
sense to make this a build option so just default to KASAN_GENERIC and
disable KASAN_SW_TAGS.

While at it, disable TEST_KASAN_MODULE in the generic 5.10 config.

Fixes build failures with KERNEL_KASAN enabled.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agokernel: 5.10: compress 7xx patch numbering
Matthew Hagan [Sat, 6 Nov 2021 13:50:27 +0000 (13:50 +0000)]
kernel: 5.10: compress 7xx patch numbering

The qca8k patch series brings the numbering to 799. This patch renames
7xx patches to create space for more backports to be added.

Signed-off-by: Matthew Hagan <mnhagan88@gmail.com>
[rename 729->719]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
3 years agoipq40xx: use zImage for Cell-C RTL30VW
Pawel Dembicki [Fri, 5 Nov 2021 15:01:54 +0000 (16:01 +0100)]
ipq40xx: use zImage for Cell-C RTL30VW

After switch to 5.10 kernel, kernel size was too high.
This patch switches Cell-C RTL30VW from uImage to zImage build.

Lzma uImage wrap is required for factory booting and it must left
untouched.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
3 years agoramips: add broken-flash-reset for HLK-7621A EvB
Wout Bertrums [Fri, 5 Nov 2021 14:09:28 +0000 (15:09 +0100)]
ramips: add broken-flash-reset for HLK-7621A EvB

This is needed because the HLK-7621 EvB has 32MB of flash,
so it will have to use 4B addressing and the
broken-flash-reset hack has to be used to be able to reboot.

Signed-off-by: Wout Bertrums <wout@wbnet.eu>
[copied github message into commit message]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
3 years agoipq40xx: 5.10: refresh config
Robert Marko [Thu, 4 Nov 2021 11:53:24 +0000 (12:53 +0100)]
ipq40xx: 5.10: refresh config

It looks like CONFIG_BLK_CMDLINE_PARSER was forgotten during the Orbi
device merge.
So lets refresh the config with it.

Signed-off-by: Robert Marko <robert.marko@sartura.hr>
3 years agox86: sort alphabetically default packages
Alberto Bursi [Wed, 3 Nov 2021 19:37:27 +0000 (20:37 +0100)]
x86: sort alphabetically default packages

sorting alphabetically default packages
and placing them on their own line.

Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
[fixed whitespaces before tab, double whitespaces]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
3 years agox86: enable kmod-tg3 on 64-bit by default
Alberto Bursi [Wed, 3 Nov 2021 19:30:56 +0000 (20:30 +0100)]
x86: enable kmod-tg3 on 64-bit by default

Gigabit ethernet adapters using BCM5719/5720 chipset
are common on servers and as easy/cheap to get as
Intel based ones.
Usually found in 2-port and 4-port cards.

Also some devices recently added to x86_64 target
like the Meraki MX100 use this chipset for 8 of
their 12 integrated ports.

Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
3 years agoramips: mt7621-dts: properly organize pcie node
Sergio Paracuellos [Wed, 5 May 2021 12:17:36 +0000 (14:17 +0200)]
ramips: mt7621-dts: properly organize pcie node

Device tree pcie node for this SoC is using different
styles in its different properties. Hence properly
unify them to be able to write a a proper yaml schema
documentation.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20210505121736.6459-11-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: add missing device_type in pcie root ports
Sergio Paracuellos [Thu, 6 May 2021 17:07:42 +0000 (19:07 +0200)]
ramips: mt7621-dts: add missing device_type in pcie root ports

According to the YAML schema 'pci-bus.yaml' the 'device_type'
property is mandatory for all pcie root ports. Hence add it.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20210506170742.28196-3-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: remove 'bug-range' property
Sergio Paracuellos [Thu, 6 May 2021 17:07:41 +0000 (19:07 +0200)]
ramips: mt7621-dts: remove 'bug-range' property

Property 'bus-range' when values are the default are
not necessary to be defined. Hence, remove all of them.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20210506170742.28196-2-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: use standard 'syscon' string
Sergio Paracuellos [Wed, 5 May 2021 13:21:54 +0000 (15:21 +0200)]
ramips: mt7621-dts: use standard 'syscon' string

Both 'memc' and 'sysc' nodes are not using 'syscon'
as a node string which is the standard one to be used.
Update both of them.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20210505132154.8263-3-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: remove ethsys node
Sergio Paracuellos [Wed, 5 May 2021 13:21:53 +0000 (15:21 +0200)]
ramips: mt7621-dts: remove ethsys node

DT 'ethsys' node is being configured as a syscon
to get access to reset and other registers in the
'mediateķ,mt7621-eth' driver. Since the 'sysc' is also
a syscon, provides the clock and also is virtually
mapped from the same physical address 0x1e000000 we
can just use 'sysc' as the phandle for the syscon in
the ethernet node. Compatible string 'mediatek,mt7621-ethsys'
of the node is not being used anywhere inside the kernel
so, this node can be safely removed.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20210505132154.8263-2-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: remove obsolete switch node
DENG Qingfang [Fri, 8 Jan 2021 02:51:55 +0000 (10:51 +0800)]
ramips: mt7621-dts: remove obsolete switch node

This was for OpenWrt's swconfig driver, which never made it upstream,
and was also superseded by MT7530 DSA driver.

Reviewed-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
Link: https://lore.kernel.org/r/20210108025155.31556-1-dqfext@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: properly define 'cpc' and 'mc' nodes
Sergio Paracuellos [Sat, 2 Oct 2021 06:07:06 +0000 (08:07 +0200)]
ramips: mt7621-dts: properly define 'cpc' and 'mc' nodes

'cpc' and 'mc' nodes correspond with the MIPS 'Cluster Power Controller'
and 'MIPS Common Device Memory Map' which are present in some MIPS related
boards. There is already bindings documentation for these two located in:
- Documentation/devicetree/bindings/power/mti,mips-cpc.yaml
- Documentation/devicetree/bindings/bus/mti,mips-cdmm.yaml

Hence, properly update compatible strings and align nodes with already
mainlined bindings documentation. Also, move their definition to a proper
place since both of them are not related with the palmbus at all.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20211002060706.30511-1-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: change some node hex addresses to lower case
Sergio Paracuellos [Sun, 17 Oct 2021 07:06:56 +0000 (09:06 +0200)]
ramips: mt7621-dts: change some node hex addresses to lower case

Hexadecimal addresses in device tree must be defined using lower case.
There are some of them that are still in upper case. Change them all.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20211017070656.12654-2-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoramips: mt7621-dts: make use of 'IRQ_TYPE_LEVEL_HIGH' instead of magic numbers
Sergio Paracuellos [Tue, 19 Oct 2021 10:29:15 +0000 (12:29 +0200)]
ramips: mt7621-dts: make use of 'IRQ_TYPE_LEVEL_HIGH' instead of magic numbers

Nodes 'gdma' and 'hsdma' are using magic number '4' in interrupts property.
Use 'IRQ_TYPE_LEVEL_HIGH' instead to align with the rest of the nodes in
the file.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20211019102915.15409-2-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoath79: WNDR3700/3800/MAC: utilize nvmem for caldata fetching
Christian Lamparter [Fri, 5 Nov 2021 21:50:45 +0000 (22:50 +0100)]
ath79: WNDR3700/3800/MAC: utilize nvmem for caldata fetching

converts the still popular WNDR3700 Series to fetch the
caldata through nvmem. As the "MAC with NVMEM" has shown,
there could pitfalls along the way.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
3 years agoath9k: nvmem for ath9k caldata
Christian Lamparter [Fri, 20 Aug 2021 22:32:31 +0000 (00:32 +0200)]
ath9k: nvmem for ath9k caldata

With "getting WIFI MAC from NVMEM" working on ath79 on 5.10,
the next logical step I think is to utilize nvmem subsystem
to also get the calibration data from there.

This will tremendously speed up the wifi bring-up, since
we no longer need the userspace helper for the simple
devices that can just load them from there.

included with this patch is a package/mac80211/refresh.

Tested on: WNDR3700v2, TP-Link Archer C7v2

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
3 years agokernel: bump 5.4 to 5.4.156
John Audia [Sun, 31 Oct 2021 11:08:40 +0000 (07:08 -0400)]
kernel: bump 5.4 to 5.4.156

Removed upstreamed:
  backport-5.4/790-v5.7-net-switchdev-do-not-propagate-bridge-updates-across.patch

All other patches automatically rebased.

Signed-off-by: John Audia <graysky@archlinux.us>
3 years agokernel: bump 5.10 to 5.10.77
Rui Salvaterra [Tue, 2 Nov 2021 21:05:27 +0000 (21:05 +0000)]
kernel: bump 5.10 to 5.10.77

Deleted (upstreamed):
bcm53xx/patches-5.10/081-v5.15-Revert-pinctrl-bcm-ns-support-updated-DT-binding-as-.patch [1]

[1] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-5.10.y&id=01c2881bb0e0a71b87ca425e1b763ac13855aa7e

Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
3 years agotools/mkimage: pass --static via PKG_CONFIG_EXTRAARGS
Andre Heider [Mon, 1 Nov 2021 06:59:21 +0000 (07:59 +0100)]
tools/mkimage: pass --static via PKG_CONFIG_EXTRAARGS

Remove the then unnecessary patch doing exactly that individually.

See also 09465d80 "u-boot.mk: always link host libraries static".

Signed-off-by: Andre Heider <a.heider@gmail.com>
3 years agoimagebuilder: fix local packages/ folder
Paul Spooren [Wed, 13 Oct 2021 01:36:33 +0000 (15:36 -1000)]
imagebuilder: fix local packages/ folder

This commit fixes commit "2999f810ff: build,IB: include kmods only in
local builds" which cause the local packages/ folder only to be added
for local builds but no longer for ImageBuilder created by the Buildbot.

The commits intention was to use remote kmods repositories rather than
storing them locally. Accidentally the entire handling of the local
`packages/` was removed.

Re-add the folder and include a README describing what it can be used
for.

Signed-off-by: Paul Spooren <mail@aparcar.org>
3 years agouqmi: update to git HEAD and improve proto handler script
Daniel Golle [Sat, 6 Nov 2021 02:46:36 +0000 (02:46 +0000)]
uqmi: update to git HEAD and improve proto handler script

 e303ba8 uqmi: update code generator
 7880de8 uqmi: sync data from libqmi project
 d647f8d uqmi: add more diagnostics commands
 6f95626 uim: add --uim-get-sim-state

Use newly introduce --uim-get-sim-state command to query PIN status
from modems which require using uim instead of dms command for that.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
3 years agobuild: add support for using prebuilt LLVM toolchain
Felix Fietkau [Thu, 4 Nov 2021 19:04:44 +0000 (20:04 +0100)]
build: add support for using prebuilt LLVM toolchain

When the prebuilt llvm toolchain is unpacked into the source dir,
it is automatically picked up and used by the build system, and eBPF
based packages can be selected

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agotools: ensure that BIN_DIR is created early
Felix Fietkau [Thu, 4 Nov 2021 18:05:16 +0000 (19:05 +0100)]
tools: ensure that BIN_DIR is created early

Fixes llvm-bpf buildbot failure

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agollvm-bpf: support creating a tarball in bin
Felix Fietkau [Thu, 4 Nov 2021 14:16:07 +0000 (15:16 +0100)]
llvm-bpf: support creating a tarball in bin

This can be used for adding the toolchain to an existing tree without having
to build it from scratch.
Enable building the toolchain + tarball by default on buildbot

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agollvm-bpf: move to staging_dir/host/llvm-bpf
Felix Fietkau [Tue, 2 Nov 2021 18:38:12 +0000 (19:38 +0100)]
llvm-bpf: move to staging_dir/host/llvm-bpf

This makes it easier to package it up for the download server

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agoinclude/cmake.mk: add support for overriding the host install prefix
Felix Fietkau [Tue, 2 Nov 2021 17:08:47 +0000 (18:08 +0100)]
include/cmake.mk: add support for overriding the host install prefix

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agoprocd: make rpcd dependency conditional
Felix Fietkau [Thu, 4 Nov 2021 15:54:29 +0000 (16:54 +0100)]
procd: make rpcd dependency conditional

Avoids building rpcd when not needed

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agobuild: fix host package unpack if git-src is used
Felix Fietkau [Thu, 4 Nov 2021 14:54:27 +0000 (15:54 +0100)]
build: fix host package unpack if git-src is used

If the package download step is skipped, it needs to be performed for the host
build instead.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agobuild: add SUBDIR_MAKE_DEBUG variable to make it easier to debug package makefiles
Felix Fietkau [Thu, 4 Nov 2021 14:53:50 +0000 (15:53 +0100)]
build: add SUBDIR_MAKE_DEBUG variable to make it easier to debug package makefiles

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agobcm4908: fix calculation of new cferam index
Sungbo Eo [Sun, 1 Aug 2021 13:17:04 +0000 (22:17 +0900)]
bcm4908: fix calculation of new cferam index

The arithmetic expansion fails when idx becomes a two digit number.
Fix this by relying on expr command.

root@OpenWrt:/# echo $(((028 + 0) % 1000))
/bin/ash: arithmetic syntax error
root@OpenWrt:/# echo $(($(expr 028 + 0) % 1000))
28

Fixes: a6a0b252baa6 ("bcm4908: add sysupgrade support")
Signed-off-by: Sungbo Eo <mans0n@gorani.run>
3 years agoomcproxy: bump to git HEAD
Stijn Tintel [Thu, 4 Nov 2021 13:04:45 +0000 (15:04 +0200)]
omcproxy: bump to git HEAD

 bfba2aa groups: use uloop_timeout_remaining64

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
3 years agoprocd: bump to git HEAD
Stijn Tintel [Thu, 4 Nov 2021 13:01:53 +0000 (15:01 +0200)]
procd: bump to git HEAD

 0ee8e73 trigger: use uloop_timeout_remaining64

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
3 years agorpcd: bump to git HEAD
Stijn Tintel [Thu, 4 Nov 2021 11:21:47 +0000 (13:21 +0200)]
rpcd: bump to git HEAD

 20bf958 session: use uloop_timeout_remaining64
 d11ffe9 session: use blobmsg_get_u64 for RPC_DUMP_EXPIRES

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
3 years agolibubox: bump to git HEAD
Stijn Tintel [Thu, 4 Nov 2021 11:18:13 +0000 (13:18 +0200)]
libubox: bump to git HEAD

 123e976 uloop: restore return type of uloop_timeout_remaining
 3344157 uloop: add uloop_timeout_remaining64
 c87d3e1 lua/uloop: use uloop_timeout_remaining64
 c86a894 uloop: deprecate uloop_timeout_remaining

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
3 years agohostapd: support qos_map_set without CONFIG_INTERWORKING
Felix Fietkau [Thu, 4 Nov 2021 10:45:18 +0000 (11:45 +0100)]
hostapd: support qos_map_set without CONFIG_INTERWORKING

This feature is useful on its own even without full interworking support

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agobcm53xx: add first 5.17 DTS changes
Rafał Miłecki [Thu, 4 Nov 2021 06:00:21 +0000 (07:00 +0100)]
bcm53xx: add first 5.17 DTS changes

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agolibubox: bump to git HEAD
Stijn Tintel [Wed, 3 Nov 2021 23:55:05 +0000 (01:55 +0200)]
libubox: bump to git HEAD

 be3dc72 uloop: avoid integer overflow in tv_diff

Fixes: FS#3943
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
3 years agoconfig: Activate SECCOMP also on MIPS 64
Hauke Mehrtens [Mon, 1 Nov 2021 13:48:22 +0000 (14:48 +0100)]
config: Activate SECCOMP also on MIPS 64

This activates SECCOMP also on mips64 and mips64el.

This was working fine in a basic test in qemu.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agotoolchain: Allow sanitizer on mips and mipsel
Hauke Mehrtens [Sun, 31 Oct 2021 16:40:06 +0000 (17:40 +0100)]
toolchain: Allow sanitizer on mips and mipsel

Support for libsanitizer on MIPS 32 and MIPSEL 32 was added with GCC 9.
MIPS 64 and ARC are still not supported.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agobinutils: Update to version 2.37
Hauke Mehrtens [Sun, 31 Oct 2021 17:25:36 +0000 (18:25 +0100)]
binutils: Update to version 2.37

This matches the version used in the toolchain.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agostrace: Update to version 5.14
Hauke Mehrtens [Sun, 31 Oct 2021 16:39:50 +0000 (17:39 +0100)]
strace: Update to version 5.14

Explicitly deactivate libselinux, otherwise we get a hard dependency to
libselinux.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agovalgrind: Activate also on MIPS 64
Hauke Mehrtens [Mon, 1 Nov 2021 13:47:02 +0000 (14:47 +0100)]
valgrind: Activate also on MIPS 64

This activates valgrind also on mips64 and mips64el.

This was working fine in a basic test in qemu.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agovalgrind: Update to version 3.18.1
Hauke Mehrtens [Sun, 31 Oct 2021 16:39:23 +0000 (17:39 +0100)]
valgrind: Update to version 3.18.1

This version has some improvements for musl.
This version works fine for me on MIPS 32 BE without MIPS16 even on musl.

The additional patch is needed to make valgrind use the correct syscall
numbers for new syscalls like clock_gettime64. The MIPS architecture
uses special syscall ranges which are different from most other systems.
The patch is pending upstream: https://bugs.kde.org/show_bug.cgi?id=444781

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agogdb: Update to version 11.1
Hauke Mehrtens [Sun, 31 Oct 2021 16:38:57 +0000 (17:38 +0100)]
gdb: Update to version 11.1

GDB 11.1 now depends on gmp.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agotoolchain: gdb: Update to version 11.1
Hauke Mehrtens [Sun, 31 Oct 2021 16:37:59 +0000 (17:37 +0100)]
toolchain: gdb: Update to version 11.1

The removed patch was already applied upstream.

gdb now mandatory depends on gmp, tell configure where to find it
explicitly. We already build gmp in the tools directory for gcc. Also
make it use mpfr and mpc as we also build both of them.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agoipq40xx: disable some devices due to kernel size
Robert Marko [Wed, 3 Nov 2021 15:45:38 +0000 (16:45 +0100)]
ipq40xx: disable some devices due to kernel size

Disable some of the ipq40xx devices due to their kernel size limitations.

These devices fail to build with kernel 5.10 and full buildbot config.

Signed-off-by: Robert Marko <robert.marko@sartura.hr>
[keep gl-b1300/gl-s1300 enabled, extend commit message]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
3 years agoipq40xx: switch to kernel 5.10
Adrian Schmutzler [Wed, 3 Nov 2021 21:02:42 +0000 (22:02 +0100)]
ipq40xx: switch to kernel 5.10

There have been enough tests and new developments require the new
kernel, so let's update it.

There is a bunch of devices that do not build anymore due to
kernel size limitations. These are disabled in a subsequent commit.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
3 years agohostapd: add wmm qos map set by default
Felix Fietkau [Wed, 3 Nov 2021 21:40:53 +0000 (22:40 +0100)]
hostapd: add wmm qos map set by default

This implements the mapping recommendations from RFC8325, with an
update from RFC8622. This ensures that DSCP marked packets are properly
sorted into WMM classes.
The map can be disabled by setting iw_qos_map_set to something invalid
like 'none'

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agovti: squash vtiv4 and vtiv6 packages into vti
Florian Eckert [Wed, 27 Oct 2021 13:54:25 +0000 (15:54 +0200)]
vti: squash vtiv4 and vtiv6 packages into vti

This change adds the same package behaviour as gre package.

Signed-off-by: Florian Eckert <fe@dev.tdt.de>
3 years agoqosify: add missing config option for nat support
Felix Fietkau [Wed, 3 Nov 2021 18:41:03 +0000 (19:41 +0100)]
qosify: add missing config option for nat support

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agoqosify: include nls.mk to avoid build error with full NLS support enabled
Felix Fietkau [Wed, 3 Nov 2021 17:54:04 +0000 (18:54 +0100)]
qosify: include nls.mk to avoid build error with full NLS support enabled

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 years agoimx: remove obsolete Kernel 5.4
Piotr Dymacz [Wed, 13 Oct 2021 12:44:51 +0000 (14:44 +0200)]
imx: remove obsolete Kernel 5.4

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoimx: update subtargets default kernel configs for 5.10
Piotr Dymacz [Thu, 8 Apr 2021 23:56:07 +0000 (01:56 +0200)]
imx: update subtargets default kernel configs for 5.10

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoimx: make kernel 5.10 default version
Piotr Dymacz [Thu, 8 Apr 2021 23:51:17 +0000 (01:51 +0200)]
imx: make kernel 5.10 default version

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoimx: introduce 'cortexa7' subtarget
Piotr Dymacz [Wed, 31 Mar 2021 07:44:36 +0000 (09:44 +0200)]
imx: introduce 'cortexa7' subtarget

This adds some essential files required by new 'cortexa7' subtarget,
dedicated for Cortex-A7 based NXP i.MX series. For now, the kernel
config-default focuses only on the i.MX 6UL family, as the following
changeset will introduce support for i.MX 6ULL based device. Support
for more platforms (e.g. i.MX 7) might be enabled later, while adding
more devices.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agouboot-imx: set BUILD_SUBTARGET to 'cortexa9'
Piotr Dymacz [Sun, 28 Mar 2021 22:48:15 +0000 (00:48 +0200)]
uboot-imx: set BUILD_SUBTARGET to 'cortexa9'

All currently supported devices belong to the imx/cortexa9 subtarget.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agouboot-envtools: move imx to imx_cortexa9
Piotr Dymacz [Wed, 13 Oct 2021 12:52:43 +0000 (14:52 +0200)]
uboot-envtools: move imx to imx_cortexa9

Subtarget-specific files under 'uboot-envtools' package are supported
since 6f3a05ebb0 ("uboot-envtools: support uci-default config also per
subtargets").

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoimx: split into arch-specific subtargets
Piotr Dymacz [Sun, 28 Mar 2021 22:33:20 +0000 (00:33 +0200)]
imx: split into arch-specific subtargets

Modern NXP i.MX series includes several different families, based on
single- or multi-core Arm Cortex-A CPUs. To be able to support more
families within a single target, we split the 'imx' in arch-specific
subtargets, starting with 'cortexa9' for the Cortex-A9 based i.MX 6,
already supported by the original 'imx6' target.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agouboot-imx6: rename to 'uboot-imx'
Piotr Dymacz [Sun, 28 Mar 2021 16:51:58 +0000 (18:51 +0200)]
uboot-imx6: rename to 'uboot-imx'

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agouboot-envtools: rename 'imx6' to 'imx'
Piotr Dymacz [Sun, 28 Mar 2021 16:11:23 +0000 (18:11 +0200)]
uboot-envtools: rename 'imx6' to 'imx'

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agokobs-ng: update dependencies after 'imx6' -> 'imx' rename
Piotr Dymacz [Sun, 28 Mar 2021 16:03:23 +0000 (18:03 +0200)]
kobs-ng: update dependencies after 'imx6' -> 'imx' rename

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agokernel: update dependencies after 'imx6' -> 'imx' rename
Piotr Dymacz [Sun, 28 Mar 2021 15:07:02 +0000 (17:07 +0200)]
kernel: update dependencies after 'imx6' -> 'imx' rename

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoimx6: rename target to 'imx'
Piotr Dymacz [Sun, 28 Mar 2021 14:56:18 +0000 (16:56 +0200)]
imx6: rename target to 'imx'

This is first step in migrating to a generic i.MX target which in the
next steps will also get divided into arch-specific subtargets.

In the result, this will make it possible to support, within a single
target, also other modern NXP i.MX families, like the i.MX 7, i.MX 8
or recently introduced i.MX 9.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoimx6: refresh and update target kernel configs
Piotr Dymacz [Sun, 28 Mar 2021 14:46:21 +0000 (16:46 +0200)]
imx6: refresh and update target kernel configs

This is a minor extension of the commit 26ae69fd03 ("imx6: refresh
kernel config with 5.10 symbols"), with correct and full disable of
the Arm Cortex-A7 based i.MX 6UL and 6UL{L,Z} families support and
re-enable of the Cortex-A9 based i.MX 6L{S,X}.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoimx6: image: fix Gateworks Ventana boot script filename
Piotr Dymacz [Sat, 26 Jun 2021 03:58:42 +0000 (05:58 +0200)]
imx6: image: fix Gateworks Ventana boot script filename

U-Boot for the Gateworks Ventana includes filename of the boot script in
the default/embedded environment (see 'include/configs/gw_ventana.h' in
the U-Boot sources).

This restores the old boot script filename ('6x_bootscript-ventana'),
making Ventana boards boot again.

Fixes: 8dba71dd33 ("imx6: image: drop BOOT_SCRIPT and fix DEVICE_NAME")
Reported-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agoath79: add support for Netgear R6100
Piotr Dymacz [Fri, 17 Sep 2021 12:52:04 +0000 (14:52 +0200)]
ath79: add support for Netgear R6100

Netgear R6100 is a dual-band Wi-Fi 5 (AC1200) router based on Qualcomm
Atheros (AR9344 + QCA9882) platform. Support for this device was first
introduced in 15f6f67d18 (ar71xx). FCC ID: PY312400225.

Specifications:

- Atheros AR9344 (560 MHz)
- 128 MB of RAM (DDR2)
- 128 MB of flash (parallel NAND)
- 2T2R 2.4 GHz Wi-Fi (AR9344)
- 2T2R 5 GHz Wi-Fi (QCA9882)
- 5x 10/100 Mbps Ethernet (AR9344)
- 4x internal antenna
- 1x USB 2.0 (GPIO-controlled power)
- 6x LED, 3x button (reset, Wi-Fi, WPS)
- UART (4-pin, 2.54 mm pitch) header on PCB
- 1x mechanical power switch
- DC jack for main power input (12 V)

WARNING: sysupgrade from older stable releases is not possible, fresh
installation (via vendor's GUI or TFTP based recovery) is required.
Reason for that is increased kernel partition size.

Installation:

Use the 'factory' image under vendor's GUI or via TFTP U-Boot recovery.
You can use the 'nmrpflash' tool at a boot time, before kernel is loaded
or start it manually by pressing the 'reset' button for ~20 seconds from
powering up the device (U-Boot will start TFTP server on 192.168.1.1,
use TFTP client to send the image).

Signed-off-by: Enrico Mioso <mrkiko.rs@gmail.com>
Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
3 years agobcm4908: add testing support for kernel 5.10
Rafał Miłecki [Wed, 3 Nov 2021 11:12:52 +0000 (12:12 +0100)]
bcm4908: add testing support for kernel 5.10

CONFIG_COMPAT_32BIT_TIME is a new symbol and has to be set to avoid:
Provide system calls for 32-bit time_t (COMPAT_32BIT_TIME) [N/y/?] (NEW)

CONFIG_RELOCATABLE needs to be enabled to make kernel start booting.
That raises a question: do we really need CONFIG_EXPERT=y ?

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agobcm4908: prepare kernel 5.10 support
Rafał Miłecki [Wed, 3 Nov 2021 08:33:48 +0000 (09:33 +0100)]
bcm4908: prepare kernel 5.10 support

It compiles but *doesn't* boot so it isn't enabled yet.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
3 years agopistachio: Take bootargs from device tree
Hauke Mehrtens [Sun, 24 Oct 2021 23:48:17 +0000 (01:48 +0200)]
pistachio: Take bootargs from device tree

The boot arguments are copied into the device tree by the boot loader
and taken from the device tree by the kernel.
The code which takes the boot arguments from the different sources was
reworked with kernel 5.5.

We have to activate CONFIG_MIPS_CMDLINE_DTB_EXTEND to take the boot
arguments from the device tree.

This makes the system boot on the board again.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agopistachio: Make kernel 5.10 patches apply
Hauke Mehrtens [Sun, 24 Oct 2021 18:59:54 +0000 (20:59 +0200)]
pistachio: Make kernel 5.10 patches apply

Make the patches apply on kernel 5.10 and refresh the patches and the
kernel configuration on top of kernel 5.10.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agopistachio: Copy kernel 5.4 patches to 5.10
Hauke Mehrtens [Sun, 24 Oct 2021 18:31:08 +0000 (20:31 +0200)]
pistachio: Copy kernel 5.4 patches to 5.10

This just copies the patches and the configuration from kernel 5.4 to
kernel 5.10.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agopistachio: Fix FIT image configuration name
Hauke Mehrtens [Sun, 24 Oct 2021 18:13:23 +0000 (20:13 +0200)]
pistachio: Fix FIT image configuration name

The pistachio U-Boot expects a default configuration with the name
config@1 in the FIT image. The default was changed in OpenWrt some
months ago.
This makes the board boot again.

Fixes: 9f714398e060 ("build: use config-1 instead of config@1 as default")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agokernel: Set some options in generic configuration
Hauke Mehrtens [Sun, 24 Oct 2021 18:42:06 +0000 (20:42 +0200)]
kernel: Set some options in generic configuration

Add CONFIG_USB_ETH and CONFIG_PWM_JZ4740 to generic kernel
configuration.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 years agostrace: remove code coverage makefile var
Rosen Penev [Wed, 20 Oct 2021 23:27:49 +0000 (16:27 -0700)]
strace: remove code coverage makefile var

It relies on a custom ax_code_coverage.m4 file included with strace.
Unfortunately, this conflicts with the one included with
autoconf-macros. Instead of creating a huge patch to fix it, just remove
the variable as code coverage is not used here.

Signed-off-by: Rosen Penev <rosenp@gmail.com>