This reverts commit c296ba834d.
According to several reports, the issues with the airtime fairness
changes are gone in current versions.
It's time to re-apply the patch now.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The ZyXEL NSA310 device is a Kirkwood based NAS:
- SoC: Marvell 88F6702 1200Mhz
- SDRAM memory: 256MB DDR2 400Mhz
- Gigabit ethernet: Realtek (over pcie)
- Flash memory: 128MB
- 1 Power button
- 1 Power LED (blue)
- 5 Status LED (green/red)
- 1 Copy/Sync button
- 1 Reset button
- 2 SATA II port (1 internal and 1 external)
- 2 USB 2.0 ports (1 front and 1 back)
- Smart fan
The stock u-boot cannot read ubi so it should be replaced with the
LEDE/OpenWRT's u-boot or with a u-boot from here
https://github.com/mibodhi/u-boot-kirkwood
This device's boot ROM supports "kwboot" tool
(in mainline u-boot, built automatically if CONFIG_KIRKWOOD is declared)
that sends an uboot image to the board over serial connection, it is very easy to unbrick.
The stock bootloader can use usb and read from FAT filesystems,
so the installation process is simple, place the uboot file on a USB flashdrive
formatted as FAT (here it is "openwrt-kirkwood-nsa310.bin", then connect TTL
to the board and write the following commands in the bootloader console:
usb reset
fatload usb 0 0x1000000 openwrt-kirkwood-nsa310.bin
nand write 0x1000000 0x00000 0x100000
reset
Now you are rebooting in the new u-boot, write this in its console to install the firmware:
usb reset
fatload usb 0 0x2000000 lede-kirkwood-nsa310b-squashfs-factory.bin
nand erase.part ubi
nand write 0x2000000 ubi 0x600000
If your firmware file is bigger than 6 MiBs you should write its size in hex
instead of 0x600000 above, or remove that number entirely (it will take a while in this case).
If you are using another uboot that can read ubi, set mtdparts like this
mtdparts=mtdparts=orion_nand:0x00c0000(uboot),0x80000(uboot_env),0x7ec0000(ubi)
And set your bootcmd to be like this
bootcmd=run setenv bootargs; ubi part ubi; ubi read 0x800000 kernel; bootm 0x800000
Then you can install the firmware as described above.
After you installed (or configured) the u-boot for booting the firmware,
write the device's mac address in the ethaddr u-boot env.
The MAC address is usually on a sticker under the device (one of the two codes is the serial),
it should begin with "107BEF" as it is assigned to ZyXEL.
write in the u-boot console (use your MAC address instead of the example)
setenv ethaddr 10:7B:EF:00:00:00
saveenv
to save the mac address in the u-boot.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
accessing the u-boot's envs on this device is required to read the mac address.
These are the envs of the new u-boot, not of the stock one.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Instead of referencing u-boot packages from device profiles and having a
-all metapackage, make the u-boot packages hidden (they don't install to
bin/ anyway), and name the files in KERNEL_BUILD_DIR appropriately
Signed-off-by: Felix Fietkau <nbd@nbd.name>
this commit allows to make a standalone u-boot for nsa310b.
While both first-stage and second-stage u-boot work fine if
installed to flash or loaded with kwboot,
I could not get stock u-boot nor bodhi's u-boot to chainload
any second stage u-boot (I also tried with dockstar's uboot
that works fine on this device if loaded with kwboot).
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Upon first invocation, the ccache program will create the required directory
hierarchy so there is no point in shipping these empty directories.
Removing those paths also avoids shipping dangling symlinks in case the
directories got linked elsewhere, e.g. into a shared global cache.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
According to some reports, -march=pentium-mmx is a better choice for
older Geode CPUs than -march=geode anyway.
Bump the minimum architecture of the legacy target from i486 to
pentium-mmx. Anything older is not worth supporting anyway.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The le64 and be64 subtargets do not share a package architecture with
any other targets, so they are pretty wasteful for a development-only
target.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
While rt288x only has a MIPS 4KEc processor, it implements the MIPS32r2
architecture just like the 24Kc, so the instruction set should be 100%
compatible.
Switching it to 24kc allows it to share the package architecture with a
lot of other targets instead of creating a special case, saving
buildbot resources.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The subtarget on which the driver still depends was removed with
dee8986b95 because it was unmaintained
for a long time.
Signed-off-by: Mathias Kresin <dev@kresin.me>
This avoids repeatedly unpacking and rebuilding packages that are
failing the build. Re-running the failing step should be much faster.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The hostapd_append_wpa_key_mgmt() procedure uses the possibly uninitialized
$ieee80211r and $ieee80211w variables in a numerical comparisation, leading
to stray "netifd: radio0 (0000): sh: out of range" errors in logread when
WPA-PSK security is enabled.
Ensure that those variables are substituted with a default value in order to
avoid emitting this (harmless) shell error.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Calling the clean target removes all .ipk files and un-stages the
package. Add a new target just for clearing the build dir and call that
one instead of the full clean target
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The libtool target package stages its files into the host staging directory
and moves the libltdl library parts from there into the target staging
directory afterwards.
By doing so, the package essentially renders the host libtool infrastructure
unusable, leading to the below error in subsequent package builds:
libtoolize: $pkgltdldir is not a directory: `.../hostpkg/share/libtool`
Prevent this problem by using a dedicated libltdl install prefix in order to
avoid overwriting and moving away preexisting files belonging to tools/libtool.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Add PROVIDES:=openvpn to the default recipe in order to let all build variants
provide a virtual openvpn package.
The advantage of this approach is that downstream packages can depend on just
"openvpn" without having to require a specific flavor.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The last two parameters passed between user space tc and kernel space
sched-cake were transposed due to a merge mistake in a parameter header
file.
As such, using a packet overhead figure was likely to set cake to wash
packet DSCP values. Similarly, the DSCP wash flag was used as an offset
to the displayed packet overhead value.
Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
