I've also killed the legacy EKH01-01 to keep things straightforward,
as this will enable us to remove all the BCF forcing code.
Approved-by: Sophronia Koilpillai
Approved-by: Lyall Beveridge
- push EKH01 -> MMx108-EKH01 change into more places
- separate SDIO from SPI images so:
- we can enable RPI's internal WiFi on SPI images
- we no longer have errors related to SPI/SDIO missing
- we can have a shared SDIO image that most people can use
- have an EKH style image which an RPI can use (i.e. no overlays)
(also leaves wifi enabled)
- all old images will upgrade to equivalent new images
- retain 'legacy' SPI+SDIO images with BCF forcing if
people prefer these, particularly since we don't
force BCF selection on the landing page
(and use would have to go to the networking config)
Approved-by: Sophronia Koilpillai
Approved-by: Evan Benn
Previously, this was done in the frontend display only (e.g. model name).
We should have just properly done the change originally.
Approved-by: Prabha Murugesan
Approved-by: Lyall Beveridge
I think these will all have enough space for this. At least ekh19
has a 256mb flash, so it's not at all constrained.
Approved-by: Sophronia Koilpillai
Approved-by: Andrew Dodd
Aside from it being a bit weird (why do we want this firmware on non
mm6108 devices?), this meant that if you didn't select a normal
firmware (e.g. mm6108/mm8108) the driver would still be included
since the tlm firmware would be there. This is confusing for
users of our fork bringing up new devices.
Approved-by: Arien Judge
Approved-by: Evan Benn
There are two downsides to all kmods:
- they increase build times when you're switching kernels
- you sometimes find odd dependency issues in upstream
which you don't want to fix
This change makes the minimal (-m) build config not use
ENABLE_ALL_KMODS. i.e. the defaults are the same, but using
'-m' the build time is significantly decreased.
We also amend the config application code so that
it's possible to override (almost) anything in
target_diffconfig.
Approved-by: Sophronia Koilpillai
Approved-by: Evan Benn
Useful for qemu virtualisation, including big endian testing.
Unfortunately, USB interface seems unstable (maybe to do with quality of PCI virtualisation?).
Approved-by: Sophronia Koilpillai
Approved-by: Evan Benn
This is particularly useful for qemu virtualisation
(but could also be used on any standard x86 system).
Approved-by: Sophronia Koilpillai
Approved-by: Evan Benn
- add the ability to override the reset button handling
- add a package which handles the 'improved' reset button handling
- make the ekh-mt76x8 targets have a reset button
- allow diag_override.sh to use existing code in diag.sh
Approved-by: Sophronia Koilpillai
Approved-by: Milad Mohtashamirad
dpp-qrcode-key combines dpp-key-recovery (private key) with the
qrcode generation (i.e. public key + mac address, previously
in netifd-morse).
Approved-by: Milad Mohtashamirad
Approved-by: Lyall Beveridge
This lets you load a kernel from userspace.
Why is this useful? Because you might want to pass modparams
to a kernel, and OpenWrt's kernel ignores modparams passed
via uboot (presumably since they often don't control the uboot).
Note that this is not necessary for an ekh01/rpi, since you
can set the modparams via the kernel command line in
/boot/config.txt
Approved-by: Andrew Dodd
Approved-by: Lyall Beveridge
With BUILD_LOG on each package stores compilation output in a separate
file. This allows us to make the jenkins console log output quieter
without losing the compilation logs.
Build like this for a general overview on tty, as well as excessive detail for
each package in the logs folder:
make -j8 V=sc | tee logs/build.log | stdbuf -oL grep -i -E "^make.*(error|[[[:digit:]]])"
Approved-by: James Haggerty
Approved-by: Lyall Beveridge
EKH19 beryl mt3000 has a uboot env which appears to be empty. It also
has a factory partition that this script can read.
Approved-by: Andrew Dodd
Approved-by: Lyall Beveridge
Now that we're getting bcf info back, this is no longer as useful.
Moreover, it also affects the regulatory info page, which IMO
should represent all available channels.
Approved-by: Andrew Dodd
Approved-by: Evan Benn
Also known as glinet spitz. The mips processor is big endian, so the
current release driver 1.14.1 will not work. I have however tested this
with the big endian patchset and it works.
The board has only 16MB flash, so I have restricted the packages.
The usb port is disabled by a gpio in the upstream dts. I have edited
that to enable the port at boot.
Approved-by: James Haggerty
Approved-by: Lyall Beveridge
