Use the new timer driver for the RTL930x devices.
Signed-off-by: Olliver Schinagl <oliver@schinagl.nl>
[remove old clock provider, select MIPS_EXTERNAL_TIMER and refresh
kernel config]
Signed-off-by: Sander Vanheule <sander@svanheule.net>
Before calling sched_clock_register(), the timer used to drive the
scheduling clock should already be enabled. Otherwise the kernel log
will show strange time jumps during, and the watchdog might not be
pinged in a timely fashion, resulting in reboots.
[ 0.160281] NET: Registered PF_NETLINK/PF_ROUTE protocol family
[ 78.104319] clocksource: Switched to clocksource realtek_otto_timer
Fixes: 3cc8011171 ("realtek: resurrect timer driver")
Signed-off-by: Sander Vanheule <sander@svanheule.net>
This adds missing HE modes to mac80211_prepare_ht_modes.
Previously mesh without wpa_supplicant would be initialized with 802.11g
/NO-HT only, as this method did not parse channel bandwidth for HE
operation.
Signed-off-by: David Bauer <mail@david-bauer.net>
Fixes: mxs: add generic subtarget (64ef920)
Adding the generic target caused the TARGET_BOOTFS_PARTSIZE to stay
hidden for these boards, crashing the FAT filesystem creation.
Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
Because this comment is followed by another comment, nothing luckily
breaks, so only a cosmetic change.
Signed-off-by: Olliver Schinagl <oliver@schinagl.nl>
Update which contains just a following fix:
* ubusd: add lookup command queuing support
Defers and continues a client's lookup command to avoid unnecessary
buffering under load.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Current job triggers based on matching of changed paths is quite
limited, so lets make it possible to additionally trigger manual CI jobs
by adding CI specific pull request build labels:
* `ci:target:x86:64` label is going to trigger CI target check jobs for
x86/64 (sub)target.
* `ci:kernel:x86:64` label is going to trigger CI kernel check jobs for
x86/64 (sub)target.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Specifications:
SOC: QCA9588 CPU 720 MHz AHB 200 MHz
Switch: AR8236
RAM: 64 MiB DDR2-600
Flash: 8 MiB
WLAN: Wi-Fi4 2.4 GHz 3*3
LAN: LAN ports *4
WAN: WAN port *1
Buttons: reset *1 + wps *1
LEDs: ethernet *5, power, wlan, wps
MAC Address:
use address source
label 70:62:b8:xx:xx:96 lan && wlan
lan 70:62:b8:xx:xx:96 mfcdata@0x35
wan 70:62:b8:xx:xx:97 mfcdata@0x6a
wlan 70:62:b8:xx:xx:96 mfcdata@0x51
Install via Web UI:
Apply factory image in the stock firmware's Web UI.
Install via Emergency Room Mode:
DIR-629 A1 will enter recovery mode when the system fails to boot or
press reset button for about 10 seconds.
First, set IP address to 192.168.0.1 and server IP to 192.168.0.10.
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.
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
RTL931x kernel builds were patched to bypass the LINKER_LOAD_ADDRESS
parameter, and hardcode it to 0x80220000. This doesn't make much sense,
since value of LINKER_LOAD_ADDRESS, load-ld, only appears to be a copy
of load-y, adjusted to the linker's taste.
Dropping the hacks for bypassing LINKER_LOAD_ADDRESS results in a kernel
that actually starts booting on an RTL9313 (Netgear MS510TXM), but
currently still hangs when the kernel switches timers.
Signed-off-by: Sander Vanheule <sander@svanheule.net>
This change consolidates Netgear EX7300 series devices into two images
corresponding to devices that share the same manufacturer firmware
image. Similar to the manufacturer firmware, the actual device model is
detected at runtime. The logic is taken from the netgear GPL dumps in a
file called generate_board_conf.sh.
Hardware details for EX7300 v2 variants
---------------------------------------
SoC: QCN5502
Flash: 16 MiB
RAM: 128 MiB
Ethernet: 1 gigabit port
Wireless 2.4GHz (currently unsupported due to lack of ath9k support):
- EX6250 / EX6400 v2 / EX6410 / EX6420: QCN5502 3x3
- EX7300 v2 / EX7320: QCN5502 4x4
Wireless 5GHz:
- EX6250: QCA9986 3x3 (detected by ath10k as QCA9984 3x3)
- EX6400 v2 / EX6410 / EX6420 / EX7300 v2 / EX7320: QCA9984 4x4
Signed-off-by: Wenli Looi <wlooi@ucalgary.ca>
We may find in a situation where due the queue an old run finish after
the last run, resulting in the containers getting overwritten with an
old version.
Limit the push-containers workflow to one concurrent run and cancel any
run in progress.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
The TP-LINK TL-ST1008F has active-high LEDs, so we need a device tree
property to express this.
Signed-off-by: Lorenz Brun <lorenz@brun.one>
[Tidy up code, restrict changes to 5.15]
Signed-off-by: Sander Vanheule <sander@svanheule.net>
On RTL931x builds, CONFIG_RTL931X was used as a stand-in for
CONFIG_NO_EXCEPT_FILL. Now that the latter is always selected for
devices in the realtek target, this hack can be removed. Resulting
device images are binary identical.
Signed-off-by: Sander Vanheule <sander@svanheule.net>
It seems like we are offsetting the KERNEL_ENTRY to +0x400, which is
also accomplished by the NO_EXCEPT_FILL configuration option.
Since this is the default for MIPS_GENERIC_KERNEL, lets push a little
bit closer to that one by doing the same thing.
Signed-off-by: Olliver Schinagl <oliver@schinagl.nl>
It appears that only a few users are using the pistachio SoC. The most
active user of the target has already approved the testing kernel and
so it is very unlikely bugs will be reported in the near future.
Therefore, the target should be directly bumped to 5.15.
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Nick Hainke <vincent@systemli.org>
Copy config and patch from kernel 5.10 to kernel 5.15.
Signed-off-by: Nick Hainke <vincent@systemli.org>
[Updated the copy]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Directly copy prebuilt tools in container instead of creating an
archieve and extracting it later in other workflows.
Update build workflow to support this new implementation.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
We can now drop the dl dir in the prebuilt tools tar as package archieve
is not a requirement anymore and won't trigger a package recompile.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Packages in general use 4 check to trigger a recompile:
- timestamp for the build_dir
- timestamp for the staging stamp dir
- depends hash for the build_dir prepared file
- presence of package archieve in dl
If host tools are prebuilt and shipped in a container or manually
installed from an archieve, it would be ideal to skip including the
package archieve and just provide the build_dir prepared files and the
staging stamp file (and the actualy prebuilt tools).
Add some logic to skip dl download for host tools if AUTOREMOVE is
selected and checks for the presence of staging dir stamp file and build
dir stamp file.
If one of these requirements are not met, the package is redownloaded
and rebuilt.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
It's possible to have prebuilt tools already extracted. Add option to
just refresh the timestamps.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Host tools path may be a symbolic link. Use -H with find to follow path
links passed from command line to find command.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Recent backport of NVMEM layout support as well as acommpanying OF changes
introduced a false #nvmem-cell-cells warning as #nvmem-cell-cells are
fully optional.
So, backport an upstream fix for this.
Fixes: 11759a5bf3 ("kernel: backport of changes & helpers")
Signed-off-by: Robert Marko <robimarko@gmail.com>
1ea5855 partname: Introduce fstools_partname_fallback_scan option
While at it also drop AUTORELEASE from PKG_RELEASE.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
This device has two sets of volumes: main ones (`kernel`, `rootfs`, etc) and
'backup' (`kernel.b`, `rootfs.b`, etc). Bootloader tries to determine which set of
volumes to use by looking at contens of `extra-para` and `extra-para.b` volumes.
These volumes contain JSON that looks like this:
```
{
"dbootFlag": "1",
"integerFlag": "1",
"fwFlag": "GOOD",
"score":1
}
```
It looks like the bootloader looks for `"fwFlag": "GOOD"` (as opposed to `BAD`)
then it compares `score` field - whichever 'good' volume has bigger score wins.
This determines which set of volumes to use to boot.
So for example if `extra-para` is good and has bigger score then `kernel`,
`rootfs`, etc volumes are used. This means bootloader needs to explain to the
kernel which volume to use for the rootfs. After looking at bootloader code with
disassembler I think it contains a bug. Relevant part of code looks something
like this:
```
if (image_id == 0) {
rootfs_volume_id = 8;
rootfs_volume_name = "rootfs";
}
else {
rootfs_volume_id = 0xf;
rootfs_volume_name = "rootfs.b";
}
sprintf(
&buffer,
0x800,
"console=ttyS0,115200 noinitrd ubi.mtd=3,2048 ubi.block=0,%s
root=/dev/ubiblock0_%d DKMGT_IMAGE_ID=%d DKMGT_IMAGE_TYPE=ubi",
rootfs_volume_name,
rootfs_volume_id,
image_id
);
```
Where `image_id == 0` if 'normal' (not '*.b' set of volumes is used).
However from device dumps we know that from the factory `rootfs.b` has id 8 and
`rootfs` has id 15.
So from above we can see that ids and names of rootfs volumes do not match. More
over - they are hardcoded in the bootloader.
Both things are problematic for OpwnWRT which completely removes volumes on
update meaning that volume ids may actually change.
So instead of relying on bootloader to provide the kernel with root device this
patch forces kernel to determine root automatically - and it defaults to
`rootfs` volume which is correct for our purposes.
Overall this makes image boot fine from flash after sysupgrade from inirams.
assuming `extra-para*` volumes make bootloader use non-'*.b' set of volumes.
Signed-off-by: Nikolay Martynov <mar.kolya@gmail.com>
Fix DT_DEBUG handling on MIPS in musl libc.
With this change gdb will load the symbol files for shared libraries on MIPS too.
This patch was taken from this thread: https://www.openwall.com/lists/musl/2022/01/09/4
Signed-off-by: Hauke Mehrtens <hmehrtens@maxlinear.com>
Byte swapping code incorrectly uses the number of AES rounds to swap expanded
AES key, while swapping only a single dword in a loop, resulting in swapped
key and partially swapped expanded keys, breaking AES encryption and
decryption on VIA Padlock hardware.
This commit correctly sets the number of swapping loops to be done.
Upstream: 2bcf8e69bd
Acked-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: ValdikSS ValdikSS <iam@valdikss.org.ru>
After commit e0d2c59ee995 ("genirq: Always limit the affinity to online
CPUs", 5.10) on Linux, the cpumask passed to irq_set_affinity of irqchip
driver is limited to online CPUs. When irq_do_set_affinity called from
otto timer driver with only one secondary CPU, that CPU is not marked as
online yet, filtered out by cpu_online_mask and fall to error path.
Then, fail to set affinity for that CPU and it leads to instability of
timer on secondary CPU(s).
At least, RTL839x system will be affected.
log:
[ 37.560020] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[ 37.638025] rcu: 1-...!: (0 ticks this GP) idle=6ac/0/0x0 softirq=0/0 fqs=1 (false positive?)
[ 37.752683] (detected by 0, t=6002 jiffies, g=-1179, q=26293)
[ 37.829510] Sending NMI from CPU 0 to CPUs 1:
[ 37.886857] NMI backtrace for cpu 1 skipped: idling at r4k_wait_irqoff+0x1c/0x24
[ 37.984801] rcu: rcu_sched kthread timer wakeup didn't happen for 5999 jiffies! g-1179 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402
[ 38.132743] rcu: Possible timer handling issue on cpu=1 timer-softirq=0
[ 38.221033] rcu: rcu_sched kthread starved for 6000 jiffies! g-1179 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402 ->cpu=1
[ 38.356336] rcu: Unless rcu_sched kthread gets sufficient CPU time, OOM is now expected behavior.
[ 38.474440] rcu: RCU grace-period kthread stack dump:
...
Replace to irq_force_affinity from irq_set_affinity and ignore
cpu_online_mask to fix the issue.
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
Tested-by: Olliver Schinagl <oliver@schinagl.nl>
