Commit 1cb8cdb ("ci: use new buildbot worker images with Debian 11")
introduced new Git version with strict rules for owner of the git
directory.
To handle this and not cause major change, just move the parsing before
the change of ownership of the openwrt directory permitting the correct
run of git fetch command with the same user that did the repository
checkout.
Fixes: 1cb8cdb ("ci: use new buildbot worker images with Debian 11") Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Commit bf8187d5dc4d ("CI: use split target and subtarget in label
workflow") didn't correctly output subtarget resulting in calling with
an empty subtarget. Fix this and correctly output generated subtarget.
Fixes: bf8187d5dc4d ("CI: use split target and subtarget in label workflow") Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
CI: use split target and subtarget in label workflow
With eecc6e48117b ("CI: rework build workflow to have split target and
subtarget directly") target and subtarget are split in 2 different
variables. Label workflow were not aligned to this change and are
currently broken.
Fix them and correctly pass split target and subtarget.
Fixes: eecc6e48117b ("CI: rework build workflow to have split target and subtarget directly") Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Due to problem with migrating from master to main as the default branch
and downstream project still requiring the master branch to be present,
we currently have for push events double CI runs, one for main and one
for master. To solve this ignore any push event to the master branch for
every workflow that react on push events.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Robert Marko [Mon, 22 May 2023 21:28:05 +0000 (23:28 +0200)]
generic: 6.1: disable CONFIG_SURFACE_PLATFORMS by default
There is no need for Surface platform drivers to be enabled by default,
especially on OpenWrt so disable CONFIG_SURFACE_PLATFORMS in the generic
config.
Tony Ambardar [Sun, 21 May 2023 09:14:37 +0000 (02:14 -0700)]
bpftools: update, split off bpftool and libbpf packages
My original bpftools package made "variant" builds of bpftool and libbpf
as a convenience, since both used the same local kernel sources with the
same versioning. This is no longer the case, since the commit below
switched to using an out-of-tree build mirror hosting repos for each.
Replace bpftools with separate bpftool and libbpf packages, each simplified
and correctly versioned. Also fix the broken libbpf ABI introduced in the
same commit. Existing build .config files are not impacted.
Fixes: 00cbf6f6ab1d ("bpftools: update to standalone bpftools + libbpf, use the latest version") Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
Robert Marko [Tue, 23 May 2023 12:55:48 +0000 (14:55 +0200)]
kernel: qca-ssdk: backport support for building as kernel module
Currently, SSDK is rather special in the sense that its not being built as
a proper out of tree module at all but rather like a userspace application
and that involves a lot of make magic which unfortunately broke with make
version 4.4 and newer.
Luckily QCA finally added a way to build SSDK as an out of tree module
and it uses the kernel buildsystem which makes it compile with make 4.4
as well.
So lets backport the support for it and switch to using it.
Petr Štetiar [Tue, 23 May 2023 14:56:09 +0000 (16:56 +0200)]
ci: push-containers: trigger job on release branching
Currently all 23.05 related CI jobs are failing as the containers are
not available, so lets fix it by pushing those containers when the
version.mk changes.
Petr Štetiar [Tue, 23 May 2023 12:17:47 +0000 (14:17 +0200)]
prereq: SetupHostCommand: fix wrong check result
Tony has reported, that CI tools job is failing for him in macOS
container due to prereq check failure for GNU `install` utility.
Michael diagnosed it and from his traces it was clear, that the issue is
caused by a wrong return value in the success check case, so lets fix it
accordingly.
Fixes: f75204036ccc ("prereq-build: allow host command symlinks to update") Reported-by: Tony Ambardar <itugrok@yahoo.com> Diagnosed-by: Michael Pratt <mcpratt@pm.me> Signed-off-by: Petr Štetiar <ynezz@true.cz>
Daniel Golle [Sun, 30 Apr 2023 16:13:33 +0000 (17:13 +0100)]
mediatek: filogic: add driver for Richtek RT5190A regulator
The Richtek RT5190A is used on the MT7988 reference board. Backport and
enable the driver on the filogic subtarget, so we can support cpufreq
on the MT7988 reference board.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Sun, 30 Apr 2023 16:41:22 +0000 (17:41 +0100)]
mediatek: add driver for built-in 2.5G Ethernet PHY
Add driver for the built-in 2.5G Ethernet PHY found in the MT7988 SoC.
To function the PHY also needs firmware files which have not yet been
published via linux-firmware.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Sun, 30 Apr 2023 13:32:03 +0000 (14:32 +0100)]
generic: mt7530: backport support for the MT7988 built-in switch
Backport commits adding support for the MT7988 built-in switch to the
mt7530 driver.
This change results in the Kconfig symbol NET_DSA_MT7530 to be extended
by NET_DSA_MT7530_MDIO (everything formally covered by NET_DSA_MT7530)
and NET_DSA_MT7530_MMIO (a new driver for the MMIO-connected built-in
switch of the MT7988 SoC).
Select NET_DSA_MT7530_MDIO for all targets previously selecting
NET_DSA_MT7530, with the exception of mediatek/filogic which also
selects NET_DSA_MT7530_MMIO.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Tony Ambardar [Fri, 5 May 2023 02:17:35 +0000 (19:17 -0700)]
kernel: enable (ARM|ARM64)_MODULE_PLTS in generic config
This allows loading modules with large memory requirements, recently needed
while testing on armvirt/32. Past forum discussions [1] and bug reports [2]
also raised this and the ipq806x target already set it in response [3].
Given this increases kernel image size by only ~1KB, is generally useful on
multi-platform kernels, and enabled by default on upstream arm32 Linux, add
it to the generic config.
The setting has similar utility on arm64, is a requirement for KASLR, and
already enabled on most OpenWrt aarch64 targets, so pull this into the
top-level generic config.
Petr Štetiar [Sun, 21 May 2023 16:43:53 +0000 (18:43 +0200)]
Revert "feeds: use git-src-full to allow Git versioning"
This partially reverts commit 7fae1e5677e9bb4979c8d4ac99be4de6955b13d0
as it should be no longer necessary to do a full clone since commit 48ed07bc0b94 ("treewide: replace AUTORELEASE with real PKG_RELEASE").
Suggested-by: Thibaut VARÈNE <hacks@slashdirt.org> Signed-off-by: Petr Štetiar <ynezz@true.cz>
kernel: add disabled symbol CONFIG_CPU_LITTLE_ENDIAN on 6.1
The identical change was previously added to the 5.15 kernel in the commit 9226f1e41948 ("kernel: disable CONFIG_CPU_LITTLE_ENDIAN in generic config").
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Robert Marko [Tue, 23 May 2023 11:29:49 +0000 (13:29 +0200)]
mac80211: backport merged version of A-MSDU mesh patch
Kernel 6.1 now has fortify_memcpy_chk() and it is causing the following
warning while trying to compile backports:
CC [M] /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/net/wireless/util.o
In file included from ./include/linux/string.h:253,
from /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/backport-include/linux/string.h:3,
from ./include/linux/bitmap.h:11,
from ./include/linux/cpumask.h:12,
from ./include/linux/smp.h:13,
from ./arch/arm64/include/asm/arch_timer.h:18,
from ./arch/arm64/include/asm/timex.h:8,
from ./include/linux/timex.h:67,
from ./include/linux/time32.h:13,
from ./include/linux/time.h:60,
from /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/backport-include/linux/time.h:3,
from ./include/linux/skbuff.h:15,
from /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/backport-include/linux/skbuff.h:3,
from ./include/linux/if_ether.h:19,
from /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/backport-include/linux/if_ether.h:3,
from ./include/linux/etherdevice.h:20,
from /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/backport-include/linux/etherdevice.h:3,
from /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/net/wireless/util.c:12:
In function 'fortify_memcpy_chk',
inlined from 'ieee80211_strip_8023_mesh_hdr' at /home/robimarko/Building/AX3600/ipq807x-5.15/build_dir/target-aarch64_cortex-a53_musl/linux-ipq807x_generic/backports-6.1.24/net/wireless/util.c:590:3:
./include/linux/fortify-string.h:404:25: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
404 | __write_overflow_field(p_size_field, size);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
This issue was fixed in the final version of
("wifi: mac80211: fix receiving A-MSDU frames on mesh interfaces") that was
merged upstream but we have a older version that is using:
memcpy(&payload.eth.h_dest, mesh_addr, 2 * ETH_ALEN);
instead of:
memcpy(&payload.eth, mesh_addr, 2 * ETH_ALEN);
So, lets just backport the merged version of patch to fix the issue.
Daniel Golle [Mon, 22 May 2023 21:58:58 +0000 (22:58 +0100)]
mediatek: mt7623: build SD card without all filesystems
Allow building SD card images without having both initramfs and squashfs
present on the card, just like it has already been done for the mt7622
and filogic subtargets.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
ipq806x: 6.1: replace ZRELADDR HACK patch with new version
With 6.1 lots have changed and the platform for Makefile.boot got
dropped. Replace the patch with a new version that is alligned to the
new implementation.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Robert Marko [Wed, 26 Oct 2022 11:27:51 +0000 (13:27 +0200)]
kernel: modules: fs: adapt for kernel 6.1
Adapt filesystem kmods for building under kernel 6.1:
* Depend on kernel not being 5.10 rather than only 5.15
* kmod-fs-9p depends on kmod-fs-netfs from 5.17 as they started using
netfs helpers
* Set new KConfig options to N
Robert Marko [Fri, 21 Oct 2022 11:08:35 +0000 (13:08 +0200)]
kernel: modules: wwan: adapt for kernel 6.1
Update the WWAN kmods for compilation under kernel 6.1:
* Depend on kernel not being 5.10 rather than only 5.15
* Enable CONFIG_WWAN_DEBUGFS as its now optional from 5.17
* Add missing symbols for new WWAN drivers to generic config
kernek: use struct_group to wipe ar8216 volatile priv data
Instead of reference vlan and do strange subtraction, use the handy
struct_group() to create a virtual struct of the same size of the
members. This permits to have a more secure memset and fix compilation
warning in 6.1 where additional checks are done.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Changes:
- Refresh mtd patches with new implementation.
- Change 191-rtc-rs5c372-let_the_alarm_to_be_used_as_wakeup_source as
uie_unsupported got dropped and we now set the bit directly.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Refresh commands, run after busybox is first built once (nothing changed
compared to 1.36.0):
cd package/utils/busybox/config/
../convert_menuconfig.pl ../../../../build_dir/target-mipsel_24kc_musl/busybox-default/busybox-1.36.1
cd ..
./convert_defaults.pl ../../../build_dir/target-mipsel_24kc_musl/busybox-default/busybox-1.36.1/.config > Config-defaults.in
Manual edits needed afterward:
* Config-defaults.in: OpenWrt config symbol IPV6 logic applied to
BUSYBOX_DEFAULT_FEATURE_IPV6
* Config-defaults.in: OpenWrt logic applied to
BUSYBOX_DEFAULT_LOGIN_SESSION_AS_CHILD (commit dc92917)
* Config-defaults.in: correct the default ports that get reset
BUSYBOX_DEFAULT_FEATURE_HTTPD_PORT_DEFAULT 80
BUSYBOX_DEFAULT_FEATURE_TELNETD_PORT_DEFAULT 23
* config/editors/Config.in: Add USE_GLIBC dependency to
BUSYBOX_CONFIG_FEATURE_VI_REGEX_SEARCH (commit f141090)
* config/shell/Config.in: change at "Options common to all shells" the conditional symbol
SHELL_ASH --> BUSYBOX_CONFIG_SHELL_ASH
(discussion in http://lists.openwrt.org/pipermail/openwrt-devel/2021-January/033140.html
Apparently our script does not see the hidden option while
prepending config options with "BUSYBOX_CONFIG_" which leads to a
missed dependency when the options are later evaluated.)
* Edit a few Config.in files by adding quotes to sourced items in
config/Config.in, config/networking/Config.in and config/util-linux/Config.in (commit 1da014f)
Tested-by: Hannu Nyman <hannu.nyman@iki.fi> Signed-off-by: Nick Hainke <vincent@systemli.org>
Pietro Ameruoso [Mon, 22 May 2023 07:52:47 +0000 (09:52 +0200)]
mediatek: add support for Zyxel EX5601-T0 router
Zyxel EX5601-T0 specifics
--------------
The operator specific firmware running on the Zyxel branded
EX5601-T0 includes U-Boot modifications affecting the OpenWrt
installation.
The router boots BL2 which than loads FIP (u-boot).
U-boot has hardcoded a command to always launch Zloader "mtd read zloader 0x46000000" and than "bootm". Bootargs are deactivated.
Zloader is the zyxel booloader which allow to dual-boot ubi or ubi2, by default access to zloader is blocked.
Too zloader checks that the firmware contains a particolar file called zyfwinfo.
Additional details regarding Zloader can be found here:
https://hack-gpon.github.io/zyxel/
https://forum.openwrt.org/t/adding-openwrt-support-for-zyxel-ex5601-t0/155914
Installation
------------
Keep in mind that openwrt can only run on the UBI partition, the openwrt firmware is not able to understand the zloader bootargs.
The procedure allows restoring the UBI partition with the Zyxel firmware and retains all the OEM functionalities.
1. Unlock Zloader (this will allow to swap manually between partitions UBI and UBI2):
- Attach a usb-ttl adapter to your computer and boot the router.
- While the router is booting at some point you will read the following: `Please press Enter to activate this console.`
- As soon as you read that press enter, type root and than press enter again (just do it, don't care about the logs scrolling).
- Most likely the router is still printing the boot log, leave it boot until it stops.
- If everything went ok you should have full root access "root@EX5601-T0:/#".
- Type the following command and press enter: "fw_setenv EngDebugFlag 0x1".
- Reboot the router.
- As soon as you read `Hit any key to stop autoboot:` press Enter.
- If everything went ok you should have the following prompt: "ZHAL>".
- You have successfully unlocked zloader access, this procedure must be done only once.
2. Check the current active partition:
- Boot the router and repeat the steps above to gain root access.
- Type the following command to check the current active image: "cat /proc/cmdline".
- If `rootubi=ubi` it means that the active partition is `mtd6`
- If `rootubi=ubi2` it means that the active partition is `mtd7`
- As mentioned earlier we need to flash openwrt into ubi/mtd6 and never overwrite ubi2/mtd7 to be able to fully roll-back.
- To activate and boot from mtd7 (ubi2) enter into ZHAL> command prompt and type the following commands:
atbt 1 # unlock write
atsw # swap boot partition
atsr # reboot the router
- After rebooting check again with "cat /proc/cmdline" that you are correctly booting from mtd7/ubi2
- If yes proceed with the installation guide. If not probably you don't have a firmware into ubi2 or you did something wrong.
3. Flashing:
- Download the sysupgrade file for the router from openwrt, than we need to add the zyfwinfo file into the sysupgrade tar.
Zloader only checks for the magic (which is a fixed value 'EXYZ') and the crc of the file itself (256bytes).
I created a script to create a valid zyfwinfo file but you can use anything that does exactly the same:
https://raw.githubusercontent.com/pameruoso/OpenWRT-Zyxel-EX5601-T0/main/gen_zyfwinfo.sh
- Add the zyfwinfo file into the sysupgrade tar.
- Enter via telnet or ssh into the router with admin credentials
- Enter the following commands to disable the firmware and model checks
"zycli fwidcheck off" and "zycli modelcheck off"
- Open the router web interface and in the update firmware page select the "restore default settings option"
- Select the sysupgrade file and click on upload.
- The router will flash and reboot itself into openwrt from UBI
4. Restoring and going back to Zyxel firmware.
- Use the ZHAL> command line to manually swap the boot parition to UBI2 with the following:
atbt 1 # unlock write
atsw # swap boot partition
atsr # reboot the router
- You will boot again the Zyxel firmware you have into UBI2 and you can flash the zyxel firmware to overwrite the UBI partition and openwrt.
Working features
----------------
3 gbit lan ports
Wifi
Zyxel partitioning for coexistance with Zloader and dual boot.
WAN SFP port (only after exporting pins 57 and 10. gpiobase411)
leds
reset button
serial interface
usb port
lan ethernet 2.5 gbit port (autosense)
wan ethernet 2.5 gbit port (autosense)
Not working
----------------
voip (missing drivers or proper zyxel platform software)
Swapping the wan ethernet/sfp xor port
----------------
The way to swap the wan port between sfp and ethernet is the following:
export the pins 57 and 10.
Pin 57 is used to probe if an sfp is present.
If pin 57 value is 0 it means that an sfp is present into the cage (cat /sys/class/gpio/gpio468/value).
If pin 57 value is 1 it means that no sfp is inserted into the cage.
In conclusion by default both 57 an 10 pins are by default 1, which means that the active port is the ethernet one.
After inserting an SFP pin 57 will become 0 and you have to manually change the value of pin 10 to 0 too.
This is totally scriptable of course.
Leds description
------------
All the leds are working out of the box but the leds managed by the 2 maxlinear phy (phy 5 lan, phy6 wan).
To activate the phy5 led (rj45 ethernet port led on the back of the router) you have to use mdio-tools.
To activate the phy6 led (led on the front of the router for 2.5gbit link) you have to use mdio-tools.
Example:
Set lan5 led to fast blink on 2500/1000, slow blink on 10/100:
mdio mdio-bus mmd 5:30 raw 0x0001 0x33FC
Set wan 2.5gbit led to constant on when wan is 2.5gbit:
mdio mdio-bus mmd 6:30 raw 0x0001 0x0080
Signed-off-by: Pietro Ameruoso <p.ameruoso@live.it>
CI: rework build workflow to have split target and subtarget directly
Instead of referring to a redundant job and ENV variables, rework build
workflow to accept and require split target and subtarget and use them
directly from inputs.
Rework each user and pass a JSON of tuple to matrix include with each
target/subtarget combination to test. Special notice this doesn't use
the github actions matrix combination feature but reference each
specific tuple of target and subtarget to test.
Just a cleanup no behaviour change intended.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
MAC Address:
use address source1 source2
label 40:9b:xx:xx:xx:3c lan && wlan u-boot,env@ethaddr
lan 40:9b:xx:xx:xx:3c devdata@0x3f $label
wan 40:9b:xx:xx:xx:3f devdata@0x8f $label + 3
wlan2g 40:9b:xx:xx:xx:3c devdata@0x5b $label
wlan5g 40:9b:xx:xx:xx:3e devdata@0x76 $label + 2
Install via Web UI:
Apply factory image in the stock firmware's Web UI.
Install via Emergency Room Mode:
DIR-859 A1 will enter recovery mode when the system fails to boot
or press reset button for about 10 seconds.
First, set computer IP to 192.168.0.5 and Gateway to 192.168.0.1.
Then we can open http://192.168.0.1 in the web browser to upload
OpenWrt factory image or stock firmware. Some modern browsers may
need to turn on compatibility mode.
Shiji Yang [Sat, 20 May 2023 12:35:51 +0000 (20:35 +0800)]
ath79: improve support for D-Link DIR-8x9 A1 series
1. Remove unnecessary new lines in the dts.
2. Remove duplicate included file "gpio.h" in the device dts.
3. Add missing button labels "reset" and "wps".
4. Unify the format of the reg properties.
5. Add u-boot environment support.
6. Reduce spi clock frequency since the max value suggested by the
chip datasheet is only 25 MHz.
7. Add seama header fixup for DIR-859 A1. Without this header fixup,
u-boot checksum for kernel will fail after the first boot.
Shiji Yang [Sun, 21 May 2023 14:51:16 +0000 (22:51 +0800)]
ramips: correct page read return value of the mt7621 nand driver
read_page() need to return maximum number of bitflips instead of the
accumulated number. Change takes from upstream mt7621 u-boot [1].
* @read_page: function to read a page according to the ECC generator
* requirements; returns maximum number of bitflips
* corrected in any single ECC step, -EIO hw error
The x86_64 UML target wants to include SSSE3 optimized
crypto code which lives under /arch/x86/crypto.
However, these are not built and this causes an error.
| ERROR: module '[...]/arch/x86/crypto/sha512-ssse3.ko' is missing.
| make[3]: *** [modules/crypto.mk:990: [...]/kmod-crypto-sha512_5.15.112-1_x86_64.ipk] Error 1
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
the UML build fails during the kernel build:
| arch/um/drivers/net_kern.c: In function 'compute_hash':
| arch/um/drivers/net_kern.c:322:1: error: the frame size of 1072 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
| 322 | }
| | ^
|cc1: all warnings being treated as errors
The compute_hash() function is added by our patch:
102-pseudo-random-mac.patch
Instead of allocating a 1024 byte buffer on the stack for the SHA1
digest input, let's allocate the data on the heap. We should be
able to do that since crypto_alloc_ahash and ahash_request_alloc
also need to allocate structures on the heap.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Rafał Miłecki [Wed, 17 May 2023 09:58:16 +0000 (11:58 +0200)]
mac80211: brcm: drop brcmfmac patch waiting for register_wiphy()
That was a workaround for OpenWrt generation of config files. This patch
was used to postpone returning from probe function until loading
firmware and calling register_wiphy().
All of that is not needed anymore thanks to the ieee80211 hotplug.d
script introduced in the commit 5f8f8a366136 ("base-files, mac80211,
broadcom-wl: wifi detection and configuration"). That takes care of
generating /etc/config/wireless entries even if wireless device appears
late in the booting process.
Michael Pratt [Sun, 22 May 2022 10:10:56 +0000 (06:10 -0400)]
prereq-build: allow host command symlinks to update
This makes the prereq stage update the symlinks
installed into staging_dir/host/bin
by rearrainging the way they are verified.
Before, seeing or installing a symlink would result in
a successful exit code, and not installing a symlink
would result is a failed exit code. However,
that is not able to account for the difference
between existing good and bad links, or whether
the link would be the same if it was reinstalled,
because the check can match the program to a different path.
Instead, let a success exit code represent
identifying an existing symlink as exactly the same
as what would be installed if it did not exist,
and let a fail exit code represent
needing to install the symlink
or not having a match to the check criteria.
The failing exit code is caught by a new second attempt
for all of the check-* targets which will then indicate
to the user that there was an update by having a success
exit code when the check is run again and the link is the same.
When there is nothing to update, the checks will run only once.
This relies on the ls command to be POSIX-conformant with long format:
"path/to/link -> target/of/link"
Also, make sure the symlink is executable, not just a file,
and the directory only needs to be created once.
Fixes: #12610 Signed-off-by: Michael Pratt <mcpratt@pm.me>
IGD is only useful when accelerating a VM guest that wants to direct
render to memory in the host's framebuffer, but since OpenWrt
typically runs on headless hardware, this serves no purpose.
Also build vfio with VFIO_NOIOMMU undefined (to get all of the code
enabled), but allow it to be enabled via boot-time modparams
settings (or at run-time via sysfs writes to
"/sys/module/vfio/parameters/enable_unsafe_noiommu_mode".
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Stan Grishin [Mon, 27 Feb 2023 00:04:33 +0000 (00:04 +0000)]
x86: base-files add support for Sophos 135r3/135r3w
The Sophos SG/XG-135 revision 3 has odd numbering of eth ports
where the WAN port (as marked on the case) is:
`eth6` and `eth0`, `eth1`, `eth2`, `eth3`, `eth5`, `eth7`, `eth8` are LAN ports.
Port `eth4` seems to be the SFP port.
Also add the missing LED definition for supported Sophos devices.
Original discussion at:
https://forum.openwrt.org/t/openwrt-on-revision-3-of-sophos-desktop-appliances/152912
projects / thirdparty / openwrt.git / log
