sys/shell: make cmds submodules and add KConfig modeling#18355
sys/shell: make cmds submodules and add KConfig modeling#18355benpicco merged 5 commits intoRIOT-OS:masterfrom
Conversation
As all of them are submodules of |
|
Still WIP? |
I bet this contains still some issues. But now may be a good point in time to push for it, as there is still plenty of time until the next release to shake out any bugs it introduces. |
I started https://github.com/RIOT-OS/RIOT/actions/runs/3088682900, so we can at least see early, which release tests might fail due to this (might be of interest to @leandrolanzieri and @maribu ;-)) |
|
Ah... the testbed is currently down due :-( So this will have to wait for a definitive answer. |
|
Testbed is up again. Let's see :-) |
|
Mh, there is a regression at least between the test run on the weekend and the run I linked above (on the merge commit for this PR). For some reason, the |
As a leftovers from RIOT-OS#18355 are still present that check for `MODULE_SHELL_COMMANDS` rather than `MODULE_SHELL_CMDS`. This updates the conditionals as needed.
As a leftovers from RIOT-OS#18355 are still present that check for `MODULE_SHELL_COMMANDS` rather than `MODULE_SHELL_CMDS`. This updates the conditionals as needed.
|
Thanks! Fix provided in #18616 |
|
this seems to have broken Detailssocat - open:/dev/ttyACM0,b115200,echo=0,raw,cs8,parenb=0,cstopb=0
make: *** [/tmp/dwq.0.2965737936981785/e884ddf95438f5dda65991941a2a27a7/makefiles/tests/tests.inc.mk:22: test] Error 1 |
|
Thanks, fix provided in #18634 |
|
Not sure, if this is caused by this PR or the general flakyness of the lorawan tests, but some spec11 tests failed since there was some output missing from |
|
Also (from the output of the same test) there might be a fix required soon to the Release-Specs repo: |
With > ifconfig
ifconfig
fIface 3 HWaddr: 00:00:00:00 Frequency: 868299987Hz RSSI: -157 BW: 125kHz SF: 7 CR: 4/5 Link: down
TX-Power: 14dBm State: SLEEP Demod margin.: 0 Num gateways.: 0
OTAA
L2-PDU:255 but after: > ifconfig
ifconfig
Iface 3 HWaddr: 00:00:00:00 Frequency: 868299987Hz RSSI: -157 Link: down
TX-Power: 14dBm State: SLEEP
OTAA
L2-PDU:255 |
|
I'd say, the problem is this RIOT/sys/shell/cmds/gnrc_netif.c Line 216 in c06335b gnrc_netif_cmd_lora does not existgnrc_netif_cmd_lora in tree due to this change anymore.
|
|
See #18648 |
Model the LoRaWAN integration to GNRC's netif command (ifconfig) as submodule of it, namely `shell_cmd_gnrc_netif_lorawan`. This should fix a regression introduced by RIOT-OS#18355
Model the LoRaWAN integration to GNRC's netif command (ifconfig) as submodule of it, namely `shell_cmd_gnrc_netif_lorawan`. This should fix a regression introduced by RIOT-OS#18355
Model the LoRaWAN integration to GNRC's netif command (ifconfig) as submodule of it, namely `shell_cmd_gnrc_netif_lorawan`. This should fix a regression introduced by RIOT-OS#18355
Contribution description
Previously
shell_commandswas a "catch-all" module that includedshell commands for each and every used module that has a shell
companion. Instead, the new
shell_cmdsmodule is now used to provideshell commands as individually selectable submodules, e.g.
cmd_gnrc_icmpv6_echonow provides the ICMPv6 echo command (a.k.a.ping).
shell_commandsis still provided as a pseudo-module thatpulls in "recommended" shell companions for the used set of modules.
The intention is, that
shell_commandsprovides the same shellcommands as before.
For a handful of shell commands individual selection was already
possible. Those modules now depend on the corresponding
cmd_%moduleand they have been deprecated.
Testing procedure
Except for when explicitly using
cmd_%modules, the machine code should not change. (I do expect changes to the debug symbols due to different path and file names of the sources.)Issues/PRs references
None