Paul Spooren [Mon, 8 Aug 2022 21:37:54 +0000 (23:37 +0200)]
CI: create Docker container containing compiled tools
Currently each Kernel compilation takes about 30 minutes of which 20
minutes are used to compile our tools. While the toolchain is downloaded
and instantly ready the tools are missing.
This commit starts uploading a Docker container including compiled tools
which are ready to use. It is automatically updated whenever any tools
are changed.
Signed-off-by: Paul Spooren <mail@aparcar.org>
Co-Developed-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Christian Marangi [Fri, 9 Sep 2022 20:53:01 +0000 (22:53 +0200)]
tools: add option BUILD_ALL_HOST_TOOLS to compile all host tools
Add option to compile all host tools even if not needed.
This can be useful to prepare a universal precompiled host tools
archive to use in another buildroot and speedup compilation.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Christian Marangi [Tue, 6 Sep 2022 14:48:37 +0000 (16:48 +0200)]
build: make find_md5 reproducible with AUTOREMOVE
While experimenting with the AUTOREMOVE option in search of a way to use
prebuilt host tools in different buildroot, it was discovered that the
md5 generated by find_md5 in depends.mk is not reproducible.
Currently the hash is generated by the path of the file in addition to
the file mod time. Out of confusion, probably, there was an idea that
such command was used on the package build_dir. Reality is that this
command is run on the package files. (Makefile, patches, src)
This is problematic because the package Makefile (for example) change at
each git clone and base the hash on the Makefile mtime doesn't really
reflect if the Makefile actually changes across a buildroot or not.
A better approach is to generate an hash of each file and then generate
an hash on the sort hash list. This way we remove the problem of git
clone setting a wrong mtime while keeping the integrity of checking if a
file changed for the package as any change will result in a different
hash.
Introduce a new kind of find_md5 function, find_md5_reproducible that
apply this new logic and limit it only with AUTOREMOVE option set to
prevent any kind of slowdown due to additional hash generation.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Rosen Penev [Fri, 30 Sep 2022 08:42:51 +0000 (01:42 -0700)]
tools/cmake: fix compilation with host libzstd-dev
cmake's find_package looks at host paths first for some reason. Switch
to using pkgconfig for the search, matching other modules.
Fixes: 3848cf458ef9 ("tools/cmake: Build without some included libs")
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Paul Spooren [Sun, 20 Mar 2022 20:02:08 +0000 (20:02 +0000)]
CI: use buildbot container for building
Instead of using a fresh Linux installation which is setup every time
use the Buildbot container which is used for our own Buildbot
infrastructure, too.
While at it also tidy up the workflow to make it more consistent with
other workflow.
Signed-off-by: Paul Spooren <mail@aparcar.org>
Co-Developed-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Bradford Zhang [Fri, 30 Sep 2022 05:28:28 +0000 (13:28 +0800)]
scripts/download.pl: add tsinghua and ustc mirrors
Add https://mirrors.tuna.tsinghua.edu.cn/ and https://mirrors.ustc.edu.cn/ mirrors into download.pl to speed up download in China.
Signed-off-by: Bradford Zhang <zyc@zyc.name>
Daniel Golle [Fri, 30 Sep 2022 12:21:16 +0000 (13:21 +0100)]
ramips: add led driver to UniFi FlexHD defaut package set
Add kmod-leds-ubnt-ledbar to the packages selected by default on the
UniFi FlexHD access point.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Fri, 30 Sep 2022 12:17:07 +0000 (13:17 +0100)]
ubnt-ledbar: make package available on other targets
As also ramips/mt7621 now has a user of the ubnt-ledbar driver, make
the package available on all targets by removing the dependency on
@TARGET_mediatek_mt7622.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Wed, 28 Sep 2022 18:37:43 +0000 (19:37 +0100)]
mediatek: add support t-phy settings from efuse on MT7986
Import patches from mtk-openwrt-feeds (MTK SDK) to support reading
t-phy settings affecting PCIe as well as USB2 and USB3 from efuse.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Thu, 29 Sep 2022 17:50:23 +0000 (18:50 +0100)]
mediatek: filogic: add efuse layout to mt7986a.dtsi
efuse is used to store board-specific settings of some of the in-SoC
peripherals. Add it to device tree, so it gets probed on boot and can
be accessed by other drivers.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Wed, 14 Sep 2022 01:15:32 +0000 (02:15 +0100)]
mediatek: add patch allowing 1-byte wide access to efuse
Allow byte-wise access to mtk-efuse as some drivers require that.
Patch imported from mtk-openwrt-feeds (MTK SDK).
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Felix Fietkau [Fri, 30 Sep 2022 10:52:35 +0000 (12:52 +0200)]
kernel: fold lzma debloat patch into the patch adding the code
Reduces the number of hack patches
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Fri, 30 Sep 2022 09:48:21 +0000 (11:48 +0200)]
kernel: move kernel image cmdline hack to the octeon target
It is the only remaining user of this hack
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Fri, 30 Sep 2022 09:42:06 +0000 (11:42 +0200)]
kernel: remove hack patch, move kirkwood specific kmods to target modules.mk
Tweaking the KCONFIG line of kmod-ata-marvell-sata makes the hack patch
unnecessary
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Fri, 30 Sep 2022 09:26:47 +0000 (11:26 +0200)]
kernel: move ubnt ledbar driver to a separate package
Simplifies the tree by removing a non-upstream kernel patch and related kconfig
symbols
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Fri, 30 Sep 2022 09:07:02 +0000 (11:07 +0200)]
kernel: remove udp tunnel kconfig selection hack
The corresponding kmod package is marked as HIDDEN and selected by all
other kernel modules that need it, so the kconfig side will be in sync
without manual selection
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Wed, 28 Sep 2022 11:57:55 +0000 (13:57 +0200)]
mac80211: fix decap offload for stations on AP_VLAN interfaces
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Rosen Penev [Sat, 24 Sep 2022 23:57:40 +0000 (16:57 -0700)]
tools/cmake: Build without some included libs
Saves a little bit of time when compiling cmake.
Added patches to fix searching liblzma and zlib. The issue is that
because pkgconfig is not used, the system libraries get used.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Rosen Penev [Sun, 25 Sep 2022 00:11:52 +0000 (17:11 -0700)]
tools/expat: build with autotools again
Allows to set expat as a dependency to cmake and save on compilation
time.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Rosen Penev [Sun, 25 Sep 2022 07:22:24 +0000 (00:22 -0700)]
tools/zlib: switch to configure script
A future commit will make tools/cmake use this.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Petr Štetiar [Wed, 28 Sep 2022 09:28:06 +0000 (11:28 +0200)]
wolfssl: fix TLSv1.3 RCE in uhttpd by using 5.5.1-stable (CVE-2022-39173)
Fixes denial of service attack and buffer overflow against TLS 1.3
servers using session ticket resumption. When built with
--enable-session-ticket and making use of TLS 1.3 server code in
wolfSSL, there is the possibility of a malicious client to craft a
malformed second ClientHello packet that causes the server to crash.
This issue is limited to when using both --enable-session-ticket and TLS
1.3 on the server side. Users with TLS 1.3 servers, and having
--enable-session-ticket, should update to the latest version of wolfSSL.
Thanks to Max at Trail of Bits for the report and "LORIA, INRIA, France"
for research on tlspuffin.
Complete release notes https://github.com/wolfSSL/wolfssl/releases/tag/v5.5.1-stable
Fixes: CVE-2022-39173
Fixes: https://github.com/openwrt/luci/issues/5962
References: https://github.com/wolfSSL/wolfssl/issues/5629
Tested-by: Kien Truong <duckientruong@gmail.com>
Reported-by: Kien Truong <duckientruong@gmail.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Petr Štetiar [Thu, 29 Sep 2022 05:50:07 +0000 (07:50 +0200)]
Revert "wolfssl: fix TLSv1.3 RCE in uhttpd by using latest 5.5.1-stable release"
This reverts commit
a596a8396b1ef23cd0eda22d9a628392e70e1e1a as I've
just discovered private email, that the issue has CVE-2022-39173
assigned so I'm going to reword the commit and push it again.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Petr Štetiar [Wed, 28 Sep 2022 09:58:22 +0000 (11:58 +0200)]
wolfssl: refresh patches
So they're tidy and apply cleanly.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Petr Štetiar [Wed, 28 Sep 2022 09:28:06 +0000 (11:28 +0200)]
wolfssl: fix TLSv1.3 RCE in uhttpd by using latest 5.5.1-stable release
Fixes denial of service attack and buffer overflow against TLS 1.3
servers using session ticket resumption. When built with
--enable-session-ticket and making use of TLS 1.3 server code in
wolfSSL, there is the possibility of a malicious client to craft a
malformed second ClientHello packet that causes the server to crash.
This issue is limited to when using both --enable-session-ticket and TLS
1.3 on the server side. Users with TLS 1.3 servers, and having
--enable-session-ticket, should update to the latest version of wolfSSL.
Thanks to Max at Trail of Bits for the report and "LORIA, INRIA, France"
for research on tlspuffin.
Complete release notes https://github.com/wolfSSL/wolfssl/releases/tag/v5.5.1-stable
Fixes: https://github.com/openwrt/luci/issues/5962
References: https://github.com/wolfSSL/wolfssl/issues/5629
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Rafał Miłecki [Thu, 29 Sep 2022 03:51:25 +0000 (05:51 +0200)]
bcm53xx: backport clk driver fix for DT nodes names
It allows dropping downstream patch renaming DT nodes.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Chris Osgood [Thu, 15 Sep 2022 12:09:38 +0000 (08:09 -0400)]
build: fix warnings from grep
Fixes build warnings when using newer versions of grep.
Signed-off-by: Chris Osgood <chris_github@functionalfuture.com>
Tested-by: Georgi Valkov <gvalkov@gmail.com>
Christian Marangi [Wed, 28 Sep 2022 12:22:49 +0000 (14:22 +0200)]
scripts/download.pl: generilize and simplify download tool check
Generilize download tool check and skip other check if a download tool
has been found.
While at it also reintroduce
c836ca84e8f641e10a8349a8f9b7432b33d6cec1
that was previously dropped with aria2c support.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Kazuhiro Ito [Tue, 30 Aug 2022 12:54:38 +0000 (21:54 +0900)]
ramips: update MAC address configuration for Buffalo WSR-1166DHP
Convert MAC address and label_mac configuration of Buffalo
WSR-1166DHP to use the generic function of OpenWrt.
Apply commit
770cfe9 for WCR-1166DS to WSR-1166DHP too.
Tested on the device and MAC address is kept before and after this
change.
Signed-off-by: Kazuhiro Ito <kzhr@d1.dion.ne.jp>
Daniel Golle [Tue, 27 Sep 2022 20:08:04 +0000 (21:08 +0100)]
kernel: mtk_ppe: use MTK_FOE_STATE_INVALID instead of *_UNBIND
Instead of dropping *fix-typo-in-__mtk_foe_entry.patch which effectively
means keeping the (also wrong) assignment of MTK_FOE_STATE_BIND, rather
use MTK_FOE_STATE_INVALID as that works well on both older (NETSYS_V1)
and newer (NETSYS_V2) MediaTek SoCs.
Suggested-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Christian Marangi [Tue, 27 Sep 2022 23:13:24 +0000 (01:13 +0200)]
realtek: rtl931x: fix missing CONFIG_COMMON_CLK_REALTEK config flag
When the realtek clock driver was introduced, CONFIG_COMMON_CLK_REALTEK
was not correctly disabled for other subtarget. Add the missing config
flag to fix compilation error on buildbot.
Fixes: 4850bd887c3a ("realtek: add RTL83XX clock driver")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Rosen Penev [Tue, 27 Sep 2022 22:35:41 +0000 (15:35 -0700)]
tools: fix firmware-utils depends
When firmware-utils was converted to use cmake, the dependency was not
updated.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Christian Marangi [Tue, 27 Sep 2022 21:17:03 +0000 (23:17 +0200)]
Revert "toolchain/gcc: use zstd from tools"
This reverts commit
e6cc3ded0709aa6c7a190c31575bb5c19e204cd2.
Require more testing as it does cause compilation error.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Christian Marangi [Tue, 27 Sep 2022 21:16:46 +0000 (23:16 +0200)]
Revert "tools/zstd: build libraries as static"
This reverts commit
e3989094b801da63b7b65b2460ea30dfeb324cd3.
Require more testing as it does cause compilation error.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Rosen Penev [Sun, 25 Sep 2022 01:42:59 +0000 (18:42 -0700)]
tools/zstd: build libraries as static
Enables to get rid of rpath hack for all users.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Nick Hainke [Sun, 28 Aug 2022 13:00:54 +0000 (15:00 +0200)]
ccache: update to 4.6.3
Release Notes:
https://ccache.dev/releasenotes.html#_ccache_4_6_3
Refresh patch:
- 100-honour-copts.patch
Signed-off-by: Nick Hainke <vincent@systemli.org>
Nick Hainke [Wed, 24 Aug 2022 08:57:35 +0000 (10:57 +0200)]
tools/ccache: update to 4.6.2
Release notes:
https://ccache.dev/releasenotes.html#_ccache_4_6_2
Signed-off-by: Nick Hainke <vincent@systemli.org>
Nick Hainke [Thu, 18 Aug 2022 05:53:00 +0000 (07:53 +0200)]
tools/ccache: update to 4.6.1
Release notes:
https://ccache.dev/releasenotes.html#_ccache_4_6_1
Signed-off-by: Nick Hainke <vincent@systemli.org>
Daniel Golle [Sun, 25 Sep 2022 15:56:45 +0000 (16:56 +0100)]
kernel: rename upstreamed patches and import fixes
Move and rename patches which were merged upstream and import follow-up
fixes for MediaTek Ethernet offloading features on MT7622 and Filogic
platforms. Remove patch
793-net-ethernet-mtk_eth_soc-fix-typo-in-__mtk_foe_entry.patch
which breaks hardware flow offloading on MT7622, it will be reverted
upstream as well.
Fixes: c93c5365c0 ("kernel: pick patches for MediaTek Ethernet from linux-next")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Manas Sambhus [Thu, 4 Aug 2022 11:24:18 +0000 (16:54 +0530)]
qos-scripts: fix trailing whitespace in config files
Signed-off-by: Manas Sambhus <manas.sambhus+github@gmail.com>
Manas Sambhus [Sat, 9 Jul 2022 06:33:43 +0000 (12:03 +0530)]
qos-scripts: replace modprobe by rmmod
modprobe -r is not available on all platforms, hence use rmmod
Signed-off-by: Manas Sambhus <manas.sambhus+github@gmail.com>
Manas Sambhus [Sat, 9 Jul 2022 06:30:25 +0000 (12:00 +0530)]
kernel: netsupport: replace insmod by modprobe
Replace insmod by modprobe in TEQL hotplug script
Signed-off-by: Manas Sambhus <manas.sambhus+github@gmail.com>
Manas Sambhus [Sat, 9 Jul 2022 06:26:09 +0000 (11:56 +0530)]
ppp: use modprobe in place of insmod
This will prevent `module is already loaded` lines from
appearing in the logs when a PPP connection is reconnecting
Signed-off-by: Manas Sambhus <manas.sambhus+github@gmail.com>
Christian Marangi [Tue, 27 Sep 2022 14:54:42 +0000 (16:54 +0200)]
include/prereq-build: fix ldconfig-stub symlink on macos and BSD OSs
Some OS may have the true bin to a different location than /bin/true.
BSD based system and macos have true on /usr/bin/true.
Fix this by checking both location and take the one available in the
system.
Reported-by: Sergey V. Lobanov <sergey@lobanov.in>
Suggested-by: Huangbin Zhan <zhanhb88@gmail.com>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Nick Hainke [Sun, 11 Sep 2022 06:15:52 +0000 (08:15 +0200)]
Makefile: fix stray \ warnings with grep-3.8
We simply grep for "/usr". So no need for "-E" or "\/". Furthermore, in
the new grep versions this creates warnings.
As written in the grep-3.8 announcement:
Regular expressions with stray backslashes now cause warnings, as
their unspecified behavior can lead to unexpected results.
For example, '\a' and 'a' are not always equivalent
<https://bugs.gnu.org/39678>.
Fixes warnings in the form of:
grep: warning: stray \ before /
Signed-off-by: Nick Hainke <vincent@systemli.org>
Rosen Penev [Sun, 25 Sep 2022 05:00:40 +0000 (22:00 -0700)]
toolchain/gdb: fix expat location
GDB is not finding tools/expat. This fixes it. Move it up with the other
tools.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Rosen Penev [Sun, 25 Sep 2022 04:46:38 +0000 (21:46 -0700)]
toolchain/gcc: use zstd from tools
pkgconfig is not used for some reason. Match other used tools.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Felix Fietkau [Tue, 27 Sep 2022 11:39:12 +0000 (13:39 +0200)]
build: fix issues with targets installed via feeds
- fix including modules.mk when a target is being replaced
- fix calling make targets from target/linux
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Christian Marangi [Mon, 26 Sep 2022 18:48:25 +0000 (20:48 +0200)]
scripts/download.pl: fix support for aria2c download tool on macos
Currently we use /dev/shm to place aria2c tmp file. This is not present
on macos. Use the openwrt tmp directory instead of the linux-only
/dev/shm to save compatibility with more os.
Fixes: d39123626931 ("download.pl: add aria2c support")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Christian Marangi [Sun, 25 Sep 2022 19:43:43 +0000 (21:43 +0200)]
scripts/download.pl: fix mirrors regression for curl and wget
With the introduction of aria2c support, curl and wget no longer try to
download the file from mirrors. Fix this regression by emptying the
remaining mirrors list only when aria2c is used.
Fixes: d39123626931 ("download.pl: add aria2c support")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Kevin Darbyshire-Bryant [Tue, 20 Sep 2022 14:16:37 +0000 (15:16 +0100)]
nftables: backport fix to interval based rules
'rule inet dscpclassify dscp_match meta l4proto { udp } th dport { 3478 }
th sport { 3478-3497, 16384-16387 } goto ct_set_ef' works with
'nft add', but not 'nft insert', the latter yields:
"BUG: unhandled op 4".
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Kevin Darbyshire-Bryant [Mon, 26 Sep 2022 08:57:43 +0000 (09:57 +0100)]
sysfsutils: Define START early in file
The luci ucode rewrite exposed the definition of START as being over 1K
from start of file. Initial versions limited the search for START &
STOP to within the 1st 1K of a file. Whilst the search has been
expanded, it doesn't do any harm to define START early in the file like
all other init scripts seen so far.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Christian Marangi [Sun, 25 Sep 2022 18:34:46 +0000 (20:34 +0200)]
Revert "mt76: fix compilation error with mt7915 and kernel >= 5.15.61"
This reverts commit
6e9613844ca1ddea7e36abc44a513c89ef0598ff.
The patch was wrong in the first place as we base everything on
backports package and the compilation error was caused by an ath11k
present downstream. (will be needed later when backports package will be
updated but not now)
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Jan Hoffmann [Sun, 25 Sep 2022 12:32:31 +0000 (14:32 +0200)]
realtek: fix RTL839x egress tag for ports >= 32
Don't overwrite AS_DPM and L2LEARNING flags when dest_port is >= 32.
Fixes: 1773264a0c6d ("realtek: correct egress frame port verification")
Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Christian Marangi [Sun, 25 Sep 2022 14:04:58 +0000 (16:04 +0200)]
mt76: fix compilation error with mt7915 and kernel >= 5.15.61
With kernel 5.15.61 the define IEEE80211_MAX_AMPDU_BUF got changed to
IEEE80211_MAX_AMPDU_BUF_HE. Add patch to fix compilation error on next
5.15 kernel versions.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Christian Marangi [Sun, 25 Sep 2022 13:54:12 +0000 (15:54 +0200)]
scripts/download.pl: fix whitespace in mirror urls and drop for
Fix whitespace in mirror urls and replace for loop with join+map logic.
Fixes: d39123626931 ("download.pl: add aria2c support")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Eneas U de Queiroz [Fri, 23 Sep 2022 11:43:59 +0000 (08:43 -0300)]
wolfssl: prefer regular libwolfssl over cpu-crypto
Rename libwolfssl-cpu-crypto to libwolfsslcpu-crypto so that the
regular libwolfssl version comes first when running:
opkg install libwolfssl
Normally, if the package name matches the opkg parameter, that package
is preferred. However, for libraries, the ABI version string is
appended to the package official name, and the short name won't match.
Failing a name match, the candidate packages are sorted in alphabetical
order, and a dash will come before any number. So in order to prefer
the original library, the dash should be removed from the alternative
library.
Fixes: c3e7d86d2b (wolfssl: add libwolfssl-cpu-crypto package)
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Eneas U de Queiroz [Fri, 23 Sep 2022 11:26:51 +0000 (08:26 -0300)]
wolfssl: ABI version shouldn't depend on benchmark
Move CONFIG_PACKAGE_libwolfssl-benchmark from the top of
PKG_CONFIG_DEPENDS to after PKG_ABI_VERSION is set.
This avoids changing the ABI version hash whether the bnechmark package
package is selected or not.
Fixes: 05df135cac (wolfssl: Rebuild when libwolfssl-benchmark gets changes)
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Bradford Zhang [Tue, 13 Sep 2022 23:54:04 +0000 (07:54 +0800)]
download.pl: add aria2c support
Use aria2c download tool by default on package download if available in
the system.
aria2c permits to use multiple mirrors and may improve download speed on
special context where servers are hard to reach.
Co-authored-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: Bradford Zhang <zyc@zyc.name>
[ fix wrong var in the script and improve commit description ]
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Nick Hainke [Sun, 11 Sep 2022 11:52:31 +0000 (13:52 +0200)]
octeon: fix imagebuilder generation by introducing generic target
The generic imagebuilder does not have a generic in the name, although
this is the default naming scheme. Use bcm53xx as template for this fix.
Before the fix:
openwrt-imagebuilder-octeon.Linux-x86_64.tar.xz
After:
openwrt-imagebuilder-octeon-generic.Linux-x86_64.tar.xz
Signed-off-by: Nick Hainke <vincent@systemli.org>
Florian Eckert [Tue, 19 Oct 2021 14:15:08 +0000 (16:15 +0200)]
x86/config-5.10: geode: add missing CONFIG_WATCHDOG_CORE option
If this module is not set to y, then I get the following compilation
error during geode build.
Package kmod-w83627hf-wdt is missing dependencies for the following libraries:
watchdog.ko
Setting the linux CONFIG_WATCHDOG_CORE to y as in all other targets
fixes this issue.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
Florian Eckert [Fri, 5 Aug 2022 09:58:26 +0000 (11:58 +0200)]
x86/config-5.15: geode: add missing CONFIG_WATCHDOG_CORE option
If this module is not set to y, then I get the following compilation
error during geode build.
Package kmod-w83627hf-wdt is missing dependencies for the following
libraries:
watchdog.ko
Setting the linux CONFIG_WATCHDOG_CORE to y as in all other targets
fixes this issue.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
Florian Eckert [Wed, 3 Aug 2022 14:22:12 +0000 (16:22 +0200)]
linux/x86: add missing CONFIG_WATCHDOG_CORE to *-wdt
The packages also need a dependency to watchdog core. This commit adds the
needed WATCHDOG_CORE dependency.
https://github.com/torvalds/linux/blob/master/drivers/watchdog/Kconfig
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
Giammarco Marzano [Wed, 14 Sep 2022 12:20:43 +0000 (14:20 +0200)]
ipq40xx: Add ZTE MF289F
It's a 4G Cat.20 router used by Vodafone Italy (called Vodafone FWA)
and Vodafone DE\T-Mobile PL (called GigaCube).
Modem is a MiniPCIe-to-USB based on Snapdragon X24,
it supports 4CA aggregation.
There are currently two hardware revisions, which
differ on the 5Ghz radio:
AT1 = QCA9984 5Ghz Radio on PCI-E bus
AT2 = IPQ4019 5Ghz Radio inside IPQ4019 like 2.4Ghz
Device specification
--------------------
SoC Type: Qualcomm IPQ4019
RAM: 256 MiB
Flash: 128 MiB SPI NAND (Winbond W25N01GV)
ROM: 2MiB SPI Flash (GD25Q16)
Wireless 2.4 GHz (IP4019): b/g/n, 2x2
Wireless 5 GHz:
(QCA9984): a/n/ac, 4x4 HW REV AT1
(IPA4019): a/n/ac, 2x2 HW REV AT2
Ethernet: 2xGbE (WAN/LAN1, LAN2)
USB ports: No
Button: 2 (Reset/WPS)
LEDs: 3 external leds: Network (white or red), Wifi, Power and 1 internal (blue)
Power: 12 VDC, 1 A
Connector type: Barrel
Bootloader: U-Boot
Installation
------------
1. Place OpenWrt initramfs image for the device on a TFTP
in the server's root. This example uses Server IP: 192.168.0.2
2. Connect serial console (115200,8n1) to serial connector
GND (which is right next to the thing with MF289F MIMO-V1.0), RX, TX
(refer to this image: https://ibb.co/31Gngpr).
3. Connect TFTP server to RJ-45 port (WAN/LAN1).
4. Stop in u-Boot (using ESC button) and run u-Boot commands:
setenv serverip 192.168.0.2
setenv ipaddr 192.168.0.1
set fdt_high 0x85000000
tftp openwrt-ipq40xx-generic-zte_mf289f-initramfs-fit-zImage.itb
bootm $loadaddr
5. Please make backup of original partitions, if you think about revert to
stock, specially mtd16 (Web UI) and mtd17 (rootFS).
Use /tmp as temporary storage and do:
WEB PARITION
--------------------------------------
cat /dev/mtd16 > /tmp/mtd16.bin
scp /tmp/mtd16.bin root@YOURSERVERIP:/
rm /tmp/mtd16.bin
ROOT PARITION
--------------------------------------
cat /dev/mtd17 > /tmp/mtd17.bin
scp /tmp/mtd17.bin root@YOURSERVERIP:/
rm /tmp/mtd17.bin
6. Login via ssh or serial and remove stock partitions
(default IP 192.168.0.1):
# this can return an error, if ubi was attached before
# or rootfs part was erased before.
ubiattach -m 17
# it could return error if rootfs part was erased before
ubirmvol /dev/ubi0 -N ubi_rootfs
# some devices doesn't have it
ubirmvol /dev/ubi0 -N ubi_rootfs_data
7. download and install image via sysupgrade -n
(either use wget/scp to copy the mf289f's squashfs-sysupgrade.bin
to the device's /tmp directory)
sysupgrade -n /tmp/openwrt-...-zte_mf289f-squashfs-sysupgrade.bin
Sometimes it could print ubi attach error, but please ignore it
if process goes forward.
Flash Layout
NAND:
mtd8:
000a0000 00020000 "fota-flag"
mtd9:
00080000 00020000 "0:ART"
mtd10:
00080000 00020000 "mac"
mtd11:
000c0000 00020000 "reserved2"
mtd12:
00400000 00020000 "cfg-param"
mtd13:
00400000 00020000 "log"
mtd14:
000a0000 00020000 "oops"
mtd15:
00500000 00020000 "reserved3"
mtd16:
00800000 00020000 "web"
mtd17:
01d00000 00020000 "rootfs"
mtd18:
01900000 00020000 "data"
mtd19:
03200000 00020000 "fota"
mtd20:
0041e000 0001f000 "kernel"
mtd21:
0101b000 0001f000 "ubi_rootfs"
SPI:
mtd0:
00040000 00010000 "0:SBL1"
mtd1:
00020000 00010000 "0:MIBIB"
mtd2:
00060000 00010000 "0:QSEE"
mtd3:
00010000 00010000 "0:CDT"
mtd4:
00010000 00010000 "0:DDRPARAMS"
mtd5:
00010000 00010000 "0:APPSBLENV"
mtd6:
000c0000 00010000 "0:APPSBL"
mtd7:
00050000 00010000 "0:reserved1"
Back to Stock (!!! need original dump taken from initramfs !!!)
-------------
1. Place mtd16.bin and mtd17.bin initramfs image
for the device on a TFTP in the server's root.
This example uses Server IP: 192.168.0.2
2. Connect serial console (115200,8n1) to serial console
connector (refer to the pin-out from above).
3. Connect TFTP server to RJ-45 port (WAN/LAN1).
4. rename mtd16.bin to web.img and mtd17.bin to root_uImage_s
5. Stop in u-Boot (using ESC button) and run u-Boot commands:
This will erase RootFS+Web:
nand erase 0x1000000 0x800000
nand erase 0x1800000 0x1D00000
This will restore RootFS:
tftpboot 0x84000000 ${dir}root_uImage_s
nand erase 0x1800000 0x1D00000
nand write $fileaddr 0x1800000 $filesize
This will restore Web Interface:
tftpboot 0x84000000 ${dir}web.img
nand erase 0x1000000 0x800000
nand write $fileaddr 0x1000000 $filesize
After first boot on stock firwmare, do a factory reset.
Push reset button for 5 seconds so all parameters will
be reverted to the one printed on label on bottom of the router
Signed-off-by: Giammarco Marzano <stich86@gmail.com>
Reviewed-by: Lech Perczak <lech.perczak@gmail.com>
(Warning: commit message did not conform to UTF-8 - hopefully fixed?,
added description of the pin-out if image goes down, reformatted
commit message to be hopefully somewhat readable on git-web,
redid some of the gpio-buttons & leds DT nodes, etc.)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
INAGAKI Hiroshi [Sun, 4 Sep 2022 12:27:11 +0000 (21:27 +0900)]
ipq40xx: add support for Sony NCP-HG100/Cellular
Sony NCP-HG100/Cellular is a IoT Gateway with 2.4/5 GHz band 11ac
(WiFi-5) wireless function, based on IPQ4019.
Specification:
- SoC : Qualcomm IPQ4019
- RAM : DDR3 512 MiB (H5TC4G63EFR)
- Flash : eMMC 4 GiB (THGBMNG5D1LBAIT)
- WLAN : 2.4/5 GHz 2T2R (IPQ4019)
- Ethernet : 10/100/1000 Mbps x2
- Transceiver : Qualcomm QCA8072
- WWAN : Telit LN940A9
- Z-Wave : Silicon Labs ZM5101
- Bluetooth : Qualcomm CSR8811
- Audio DAC : Realtek ALC5629
- Audio Amp. : Realtek ALC1304
- Voice Input Processor : Conexant CX20924
- Micro Controller Unit : Nuvoton MINI54FDE
- RGB LED, Fan, Temp. sensors
- Touch Sensor : Cypress CY8C4014LQI
- RGB LED driver : TI LP55231 (2x)
- LEDs/Keys : 11x, 6x
- UART : through-hole on PCB
- J1: 3.3V, TX, RX, GND from tri-angle marking
- 115200n8
- Power : 12 VDC, 2.5 A
Flash instruction using initramfs image:
1. Prepare TFTP server with the IP address 192.168.132.100 and place the
initramfs image to TFTP directory with the name "
C0A88401.img"
2. Boot NCP-HG100/Cellular and interrupt after the message
"Hit any key to stop autoboot: 2"
3. Perform the following commands and set bootcmd to allow booting from
eMMC
setenv bootcmd "mmc read 0x84000000 0x2e22 0x4000 && bootm 0x84000000"
saveenv
4. Perform the following command to load/boot the OpenWrt initramfs image
tftpboot && bootm
5. On the initramfs image, perform sysupgrade with the sysupgrade image
(if needed, backup eMMC partitions by dd command and download to
other place before performing sysupgrade)
6. Wait for ~120 seconds to complete flashing
Known issues:
- There are no drivers for audio-related chips/functions in Linux Kernel
and OpenWrt, they cannot be used.
- There is no driver for MINI54FDE Micro-Controller Unit, customized for
this device by the firmware in the MCU. This chip controls the
following functions, but they cannot be controlled in OpenWrt.
- RGB LED
- Fan
this fan is controlled automatically by MCU by default, without
driver
- Thermal Sensors (2x)
- Currently, there is no driver or tool for CY8C4014LQI and cannot be
controlled. It cannot be exited from "booting mode" and moved to "normal
op mode" after booting. And also, the 4x buttons (mic mute, vol down,
vol up, alexa trigger) connected to the IC cannot be controlled.
- it can be exited from "booting mode" by installing and executing
i2cset command:
opkg update
opkg install i2c-tools
i2cset -y 1 0x14 0xf 1
- There is a connection issue on the control by uqmi for the WWAN module.
But modemmanager can be used without any issues and the use of it is
recommended.
- With the F2FS format, too many errors are reported on erasing eMMC
partition "rootfs_data" while booting:
[ 1.360270] sdhci: Secure Digital Host Controller Interface driver
[ 1.363636] sdhci: Copyright(c) Pierre Ossman
[ 1.369730] sdhci-pltfm: SDHCI platform and OF driver helper
[ 1.374729] sdhci_msm
7824900.sdhci: Got CD GPIO
...
[ 1.413552] mmc0: SDHCI controller on
7824900.sdhci [
7824900.sdhci] using ADMA 64-bit
[ 1.528325] mmc0: new HS200 MMC card at address 0001
[ 1.530627] mmcblk0: mmc0:0001 004GA0 3.69 GiB
[ 1.533530] mmcblk0boot0: mmc0:0001 004GA0 partition 1 2.00 MiB
[ 1.537831] mmcblk0boot1: mmc0:0001 004GA0 partition 2 2.00 MiB
[ 1.542918] mmcblk0rpmb: mmc0:0001 004GA0 partition 3 512 KiB, chardev (247:0)
[ 1.550323] Alternate GPT is invalid, using primary GPT.
[ 1.561669] mmcblk0: p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 p13 p14 p15 p16 p17
...
[ 8.841400] mount_root: loading kmods from internal overlay
[ 8.860241] kmodloader: loading kernel modules from //etc/modules-boot.d/*
[ 8.863746] kmodloader: done loading kernel modules from //etc/modules-boot.d/*
[ 9.240465] block: attempting to load /etc/config/fstab
[ 9.246722] block: unable to load configuration (fstab: Entry not found)
[ 9.246863] block: no usable configuration
[ 9.254883] mount_root: overlay filesystem in /dev/mmcblk0p17 has not been formatted yet
[ 9.438915] urandom_read: 5 callbacks suppressed
[ 9.438924] random: mkfs.f2fs: uninitialized urandom read (16 bytes read)
[ 12.243332] mmc_erase: erase error -110, status 0x800
[ 12.246638] mmc0: cache flush error -110
[ 15.134585] mmc_erase: erase error -110, status 0x800
[ 15.135891] mmc_erase: group start error -110, status 0x0
[ 15.139850] mmc_erase: group start error -110, status 0x0
...(too many the same errors)...
[ 17.350811] mmc_erase: group start error -110, status 0x0
[ 17.356197] mmc_erase: group start error -110, status 0x0
[ 17.439498] sdhci_msm
7824900.sdhci: Card stuck in wrong state! card_busy_detect status: 0xe00
[ 17.446910] mmc0: tuning execution failed: -5
[ 17.447111] mmc0: cache flush error -110
[ 18.012440] F2FS-fs (mmcblk0p17): Found nat_bits in checkpoint
[ 18.062652] F2FS-fs (mmcblk0p17): Mounted with checkpoint version =
428fa16b
[ 18.198691] block: attempting to load /etc/config/fstab
[ 18.198972] block: unable to load configuration (fstab: Entry not found)
[ 18.203029] block: no usable configuration
[ 18.211371] mount_root: overlay filesystem has not been fully initialized yet
[ 18.214487] mount_root: switching to f2fs overlay
So, this support uses ext4 format instead which has no errors.
Note:
- The primary uart is shared for debug console and Z-Wave chip. The
function is switched by GPIO15 (Linux: 427).
value:
1: debug console
0: Z-Wave
- NCP-HG100/Cellular has 2x os-image pairs in eMMC.
- 0:HLOS, rootfs
- 0:HLOS_1, rootfs_1
In OpenWrt, the first image pair is used.
- "bootipq" command in U-Boot requires authentication with signed-image
by default. To boot unsigned image of OpenWrt, use "mmc read" and
"bootm" command instead.
- This support is for "Cellular" variant of NCP-HG100 and not tested on
"WLAN" (non-cellular) variant.
- The board files of ipq-wifi may also be used in "WLAN" variant of
NCP-HG100, but unconfirmed and add files as for "Cellular" variant.
- "NET" LED is used to indicate WWAN status in stock firmware.
- There is no MAC address information in the label on the case, use the
address included in UUID in the label as "label-MAC" instead.
- The "CLOUD" LEDs are partially used for indication of system status in
stock firmware, use they as status LEDs in OpenWrt instead of RGB LED
connected to the MCU.
MAC addresses:
LAN : 5C:FF:35:**:**:ED (ART, 0x6 (hex))
WAN : 5C:FF:35:**:**:EF (ART, 0x0 (hex))
2.4 GHz: 5C:FF:35:**:**:ED (ART, 0x1006 (hex))
5 GHz : 5C:FF:35:**:**:EE (ART, 0x5006 (hex))
partition layout in eMMC (by fdisk, GPT):
Disk /dev/mmcblk0:
7733248 sectors, 3776M
Logical sector size: 512
Disk identifier (GUID): ****
Partition table holds up to 20 entries
First usable sector is 34, last usable sector is
7634910
Number Start (sector) End (sector) Size Name
1 34 1057 512K 0:SBL1
2 1058 2081 512K 0:BOOTCONFIG
3 2082 3105 512K 0:QSEE
4 3106 4129 512K 0:QSEE_1
5 4130 4641 256K 0:CDT
6 4642 5153 256K 0:CDT_1
7 5154 6177 512K 0:BOOTCONFIG1
8 6178 6689 256K 0:APPSBLENV
9 6690 8737 1024K 0:APPSBL
10 8738 10785 1024K 0:APPSBL_1
11 10786 11297 256K 0:ART
12 11298 11809 256K 0:HSEE
13 11810 28193 8192K 0:HLOS
14 28194 44577 8192K 0:HLOS_1
15 44578 306721 128M rootfs
16 306722 568865 128M rootfs_1
17 568866
3958065 1654M rootfs_data
[initial work]
Signed-off-by: Iwao Yuki <dev.clef@gmail.com>
Co-developed-by: Iwao Yuki <dev.clef@gmail.com>
[adjustments, cleanups, commit message, sending patch]
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
(dropped clk_unused_ignore, dropped 901-* patches, renamed
key nodes, changed LEDs chan/labels to match func-en, made
:net -> (w)wan leds)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Tomasz Maciej Nowak [Fri, 23 Sep 2022 12:48:12 +0000 (14:48 +0200)]
ipq40xx: base-files: properly close case
Fix this occurrence during boot:
/bin/board_detect: line 10: Unsupported: not found
Fixes: 80baffd2aa91 (" ipq40xx: add support for Pakedge WR-1")
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Tomasz Maciej Nowak [Fri, 23 Sep 2022 12:47:39 +0000 (14:47 +0200)]
ipq40xx: luma_wrtq-329acn: differentiate ethernet mac addresses
On OEM firmware both addresses for In and Out ports are different. Set
them as such also in OpenWrt.
Fixes: e24635710c7e (" ipq40xx: add support for Luma Home WRTQ-329ACN")
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Christian Lamparter [Fri, 16 Sep 2022 20:32:58 +0000 (22:32 +0200)]
apm821xx: make 5.15 kernel default
5.15 has been available for some time for testing.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Fri, 16 Sep 2022 22:09:27 +0000 (00:09 +0200)]
apm821xx: refresh 5.15 config
use the toolchain's default CPU (464fp) as the CPU option.
This fixes a CPU selection prompt which shows up now.
CPU selection
> 1. Generic 32 bits powerpc (POWERPC_CPU) (NEW)
2. Rely on the toolchain's implicit default CPU (TOOLCHAIN_DEFAULT_CPU) (NEW)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Kasparas Elzbutas [Mon, 18 Jul 2022 07:05:33 +0000 (10:05 +0300)]
ipq40xx: fix RUTX10 Wi-Fi woes
This partially reverts:
commit
cfc13c44595d ("ipq40xx: utilize nvmem-cells for macs & (pre-)calibration data")
U-Boot on these devices mangles the device tree,
so nvmem-cell type calibration doesn't work.
Fixes: cfc13c44595d ("ipq40xx: utilize nvmem-cells for macs & (pre-)calibration data")
Signed-off-by: Kasparas Elzbutas <elzkas@gmail.com>
(added reference to commit, rewrote commit message)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Mon, 22 Aug 2022 08:35:28 +0000 (10:35 +0200)]
firmware: intel-microcode: update to
20220809
Debian's changelog by Henrique de Moraes Holschuh <hmh@debian.org>:
* New upstream microcode datafile
20220809
* Fixes INTEL-SA-00657, CVE-2022-21233
Stale data from APIC leaks SGX memory (AEPIC leak)
* Fixes unspecified errata (functional issues) on Xeon Scalable
* Updated Microcodes:
sig 0x00050653, pf_mask 0x97, 2022-03-14, rev 0x100015e, size 34816
sig 0x00050654, pf_mask 0xb7, 2022-03-08, rev 0x2006e05, size 44032
sig 0x000606a6, pf_mask 0x87, 2022-04-07, rev 0xd000375, size 293888
sig 0x000706a1, pf_mask 0x01, 2022-03-23, rev 0x003c, size 75776
sig 0x000706a8, pf_mask 0x01, 2022-03-23, rev 0x0020, size 75776
sig 0x000706e5, pf_mask 0x80, 2022-03-17, rev 0x00b2, size 112640
sig 0x000806c2, pf_mask 0xc2, 2022-03-19, rev 0x0028, size 97280
sig 0x000806d1, pf_mask 0xc2, 2022-03-28, rev 0x0040, size 102400
sig 0x00090672, pf_mask 0x03, 2022-06-07, rev 0x0022, size 216064
sig 0x00090675, pf_mask 0x03, 2022-06-07, rev 0x0022, size 216064
sig 0x000906a3, pf_mask 0x80, 2022-06-15, rev 0x0421, size 216064
sig 0x000906a4, pf_mask 0x80, 2022-06-15, rev 0x0421, size 216064
sig 0x000a0671, pf_mask 0x02, 2022-03-17, rev 0x0054, size 103424
sig 0x000b06f2, pf_mask 0x03, 2022-06-07, rev 0x0022, size 216064
sig 0x000b06f5, pf_mask 0x03, 2022-06-07, rev 0x0022, size 216064
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Sat, 6 Aug 2022 17:44:02 +0000 (19:44 +0200)]
uml: add Kernel 5.15 support via testing
Add the latest default Kernel for testing.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Sat, 24 Sep 2022 21:42:42 +0000 (23:42 +0200)]
uml: fix 5.10 build
the 5.10 uml build currently breaks with:
/usr/bin/ld: arch/um/os-Linux/signal.o: in function `sigusr1_handler':
arch/um/os-Linux/signal.c:141: undefined reference to `uml_pm_wake'
But there's an upstream fix for this. Backport the fix
for now but also let upstream know so it finds its way
through the -stable releases.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Sat, 24 Sep 2022 17:57:26 +0000 (19:57 +0200)]
uml: headers are not exportable
When building UML this showed up as of late (2022-09-01):
| Makefile:1335: *** Headers not exportable for UML. Stop.
This message popped up because of an upstream patch that
fixed this error not showing up sooner. For more information see:
<https://lore.kernel.org/lkml/
20220901011252.550830-1-masahiroy@kernel.org/T/>
Now, I don't think we can just skip the kernel headers.
Some userspace applications are going to need these. So,
The next best alternative I can think of, is to go with
the headers by the host arch the UML is compiled for.
(Technically, ARCH= should be safe for all other targets
as well... But let's not poke a hornet's nest)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Alberto Martinez-Alvarez [Fri, 5 Aug 2022 21:10:54 +0000 (07:10 +1000)]
mac80211: fix parameter reading for AC_BE tx bursting
The "tx_burst" option which should control the value was
expecting more of a list and hence tx_queue_data2_burst
value wasn't updated.
Yes, it would make sense to have a list for this, the
existing code only updates tx_queue_data2_burst and
not the other tx_queue_data[0134]_burst values.
Signed-off-by: Alberto Martinez-Alvarez <amteza@gmail.com>
(formatted commit message, wrote extra information into commit,
moved tx_burst to existing json_get_vars)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Hauke Mehrtens [Sat, 24 Sep 2022 11:33:37 +0000 (13:33 +0200)]
kernel: Move some IOMMU options to generic
This adds some missing IOMMU related options for x86/64 and moves some
of them to generic for all targets.
On x86 IOMMU_DEFAULT_DMA_LAZY is used by default, on all other platforms
IOMMU_DEFAULT_DMA_STRICT is the default. we just follow the default
kernel configuration here.
Fixes: 8fea4a102ccd ("x86/64: enable IOMMU support")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Nick Hainke [Fri, 16 Sep 2022 10:03:19 +0000 (12:03 +0200)]
ath79: switch to 5.15 as default kernel
The 5.15 kernel has new interesting features like MGLRU. Most of the
targets already have added support for testing kernel 5.15 since April
2022. Set 5.15 as default for all subtargets.
Testing support was added here:
-
ae6bfb7d67c1 ("ath79: tiny: add 5.15 support for tiny subtarget")
-
9a0155bc4fa3 ("ath79: add 5.15 support for generic subtarget")
-
5af9aafabbc0 ("ath79: mikrotik: add 5.15 support for mikrotik subtarget")
-
f3fa68e5153b ("ath79: nand: add 5.15 support for nand subtarget")
Tested on:
- Nanostation M5 XM (tiny)
- TP-Link EAP-225 Outdoor (generic)
- TP-Link CPE210 (generic)
Signed-off-by: Nick Hainke <vincent@systemli.org>
Will Moss [Wed, 3 Aug 2022 13:56:28 +0000 (13:56 +0000)]
kernel: fix mac-address-increment on driver probe defer
Fixes situations where MAC address gets incremented multiple times
if device initialization fails at first and then is deferred.
Fixes: d284e6ef0f06 ("treewide: convert mtd-mac-address-increment* to generic implementation")
Signed-off-by: Will Moss <willormos@gmail.com>
Felix Fietkau [Fri, 23 Sep 2022 16:11:04 +0000 (18:11 +0200)]
mac80211: remove broken CFG80211_WEXT select from the r8723bs driver
Fixes a build error
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Fri, 23 Sep 2022 10:04:12 +0000 (12:04 +0200)]
mac80211: mark lib80211 as hidden
Prevents build errors by ensuring that it is only selected when a wext based
driver that needs it is also selected
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Christian Marangi [Fri, 23 Sep 2022 00:04:36 +0000 (02:04 +0200)]
generic: 5.15: add missing PAGE_POOL_STATS config
Add new PAGE_POOL_STATS config backported for the mtk offload patches.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Daniel Golle [Mon, 29 Aug 2022 06:54:41 +0000 (08:54 +0200)]
kernel: pick patches for MediaTek Ethernet from linux-next
Pick patches with several fixes and improvements, preparation for
upcoming WED (TX) [1] as well as basic XDP support [2] with MediaTek's
Filogic SoCs to the mtk_eth_soc driver.
Also pick follow-up patch fixing Ethernet on MT7621 [3].
Tested on Bananapi BPi-R3 (MT7986), Bananapi BPi-R64 (MT7622),
Bananapi BPi-R2 (MT7623), MikroTik RouterBoard M11G (MT7621).
[1]: https://patchwork.kernel.org/project/netdevbpf/list/?series=662108&state=*
[2]: https://patchwork.kernel.org/project/netdevbpf/list/?series=675368&state=*
(the first part of the series adding wed nodes to mt7986a.dtsi was
applied to the copy of mt7986a.dtsi in our tree)
[3]: https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/commit/?id=
5e69163d3b9931098922b3fc2f8e786af8c1f37e
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Michael Pratt [Mon, 5 Sep 2022 20:30:00 +0000 (16:30 -0400)]
prism54-firmware: set DL_DIR to package specific subdirectory
This package downloads raw files
which have names that are not corresponding to
the name and version of the package
as it is defined in the Makefile.
Use the option DL_SUBDIR to set the DL_DIR
to be a subdirectory named with
PKG_NAME and PKG_RELEASE
to better organize the downloads.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Mon, 5 Sep 2022 20:25:08 +0000 (16:25 -0400)]
b43legacy-firmware: set DL_DIR to package specific subdirectory
This package downloads raw files
which have names that are not corresponding to
the name and version of the package
as it is defined in the Makefile.
Use the option DL_SUBDIR to set the DL_DIR
to be a subdirectory named with
PKG_NAME and PKG_VERSION
to better organize the downloads.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Mon, 5 Sep 2022 20:22:08 +0000 (16:22 -0400)]
ath10k-ct-firmware: set DL_DIR to package specific subdirectory
This package downloads raw files
which have names that are not corresponding to
the name and version of the package
as it is defined in the Makefile.
Use the option DL_SUBDIR to set the DL_DIR
to be a subdirectory named with
PKG_NAME and PKG_VERSION
to better organize the downloads.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Mon, 5 Sep 2022 20:00:42 +0000 (16:00 -0400)]
acx-mac80211: set DL_DIR to package specific subdirectory
This package downloads raw files
which have names that are not corresponding to
the name and version of the package
as it is defined in the Makefile.
Use the option DL_SUBDIR to set the DL_DIR
to be a subdirectory named with
PKG_NAME and PKG_SOURCE_DATE
to better organize the downloads.
Define PKG_VERSION here
using PKG_SOURCE_DATE.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Mon, 19 Sep 2022 20:39:34 +0000 (16:39 -0400)]
scripts/dl_cleanup: add support for subdirectories
Allow comparing subdirectories exactly like files.
Handle a corner case where the new subdirectory
has the same tarball inside of it
as the one that was downloaded
before a subdirectory for that package was established.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Wed, 3 Aug 2022 17:01:51 +0000 (13:01 -0400)]
rules.mk: add option to download files into a subdirectory
Some packages download a single raw file or set of raw files
instead of an archive.
The filenames of these download files have names and version numbers
that do not directly correspond to the name and version of the package
as it is defined in its Makefile.
To better organize downloaded files, this adds the option of
creating a subdirectory in the download directory for these files
that can be named with package specific variables.
DL_DIR is now a recursively expanded variable
since the value now depends on whether DL_SUBDIR is defined.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Olliver Schinagl [Wed, 21 Sep 2022 15:57:53 +0000 (17:57 +0200)]
base-files: Actually set default name
The currently used shell expansion doesn't seem to exist [0] and also
does not work. This surely was not intended, so lets allow default
naming to actually work.
[0]: https://pubs.opengroup.org/onlinepubs/
9699919799/utilities/V3_chap02.html
Fixes: be09c5a3cd65 ("base-files: add board.d support for bridge device")
Signed-off-by: Olliver Schinagl <oliver@schinagl.nl>
Felix Fietkau [Thu, 22 Sep 2022 13:23:54 +0000 (15:23 +0200)]
iwinfo: update to the latest version
46f04f3808e8 devices: add MediaTek MT7986 WiSoC
b3e08c8b5a8f ops: make support for wireless extensions optional
1f695d9c7f82 nl80211: allow phy names that don't start with 'phy'
b7f9f06e1594 nl80211: fix phy/netdev index lookup
4a43b0d40ba5 nl80211: look up the phy name instead of assuming name == phy<idx>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 22 Sep 2022 12:01:52 +0000 (14:01 +0200)]
hostapd: add ubus notification on sta authorized
Also include the station auth_type in the ubus and log message in order
to detect, if clients used FT or FILS to associate
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 22 Sep 2022 11:42:40 +0000 (13:42 +0200)]
mac80211: disable drivers that rely on wireless extensions
They are unmaintained and don't work properly with current wifi scripts
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 22 Sep 2022 11:34:28 +0000 (13:34 +0200)]
kernel: disable wireless extensions only when needed
They are only needed by a few very old drivers
Signed-off-by: Felix Fietkau <nbd@nbd.name>
John Audia [Tue, 20 Sep 2022 19:34:17 +0000 (15:34 -0400)]
kernel: bump 5.15 to 5.15.69
All patches automatically rebased
Build system: x86_64
Build-tested: bcm2711/RPi4B, mt7622/RT3200
Run-tested: bcm2711/RPi4B, mt7622/RT3200
Signed-off-by: John Audia <therealgraysky@proton.me>
John Audia [Wed, 21 Sep 2022 11:52:34 +0000 (07:52 -0400)]
kernel: bump 5.10 to 5.10.144
All patches automatically rebased.
Signed-off-by: John Audia <therealgraysky@proton.me>
Hauke Mehrtens [Wed, 21 Sep 2022 15:20:37 +0000 (17:20 +0200)]
kernel: Reorder kernel configuration
This was done by running this:
./scripts/kconfig.pl '+' target/linux/generic/config-5.15 /dev/null > target/linux/generic/config-5.15-new
mv target/linux/generic/config-5.15-new target/linux/generic/config-5.15
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Wed, 21 Sep 2022 13:18:33 +0000 (15:18 +0200)]
kernel: Deactivate CONFIG_DEFAULT_FQ_PIE by default
When building OpenWrt with CONFIG_ALL_KMODS the kernel build will ask
for CONFIG_DEFAULT_FQ_PIE option. This deactivates it by default.
Fixes: c3e4a0d99b97 ("kernel: netsupport: Add FQ-PIE as an optional sched kmod and extract PIE")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Nicola Corna [Sat, 20 Aug 2022 21:41:59 +0000 (21:41 +0000)]
kernel: add VFIO kernel packages
Add kmod-vfio and kmod-vfio-pci packages.
Signed-off-by: Nicola Corna <nicola@corna.info>
Nicola Corna [Fri, 19 Aug 2022 19:34:01 +0000 (19:34 +0000)]
x86/64: enable IOMMU support
Enable IOMMU support for Intel and AMD x86 platforms. With this, when the
vfio module is present, physical PCI devices can be passed to VMs, for
example with `qemu-system-x86_64 -device vfio-pci,host=05:00.0 ...`.
IOMMU support increases the kernel size by a small amount (~370KB, from
5239840 B to
5611200 B, a ~7% increase in size).
Signed-off-by: Nicola Corna <nicola@corna.info>
Tomasz Maciej Nowak [Sun, 11 Sep 2022 16:14:17 +0000 (18:14 +0200)]
arm-trusted-firmware-mvebu: stop cluttering Image Builder
All contents of staging_dir/image are included in Image Builder (IB) in
case some binary needs to be included in final image. But in case of
this package, all sources are stored there and those clutter the final
tarball of IB for no reason. Those sources are not used during image
creation and are just dead weight. To put it in perspective, the IB for
21.02.0 is 158 MiB, 22.03.0-rc6 is 366 MiB and snapshot is over 620 MiB!
To fix it, put them in package build directory, so they won't end up
included in IB tarball.
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Reviewed-by: Andre Heider <a.heider@gmail.com>
Rafał Miłecki [Wed, 21 Sep 2022 05:53:52 +0000 (07:53 +0200)]
bcm4908: fix -EPROBE_DEFER support in bcm4908_enet
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>