franoosh has quit [Read error: Connection reset by peer]
_whitelogger has joined #linux-rockchip
<naoki>
By the way, how can I see the label property of adc-keys/gpio-keys?
matthias_bgg has joined #linux-rockchip
<qschulz>
naoki: does evtest report the label as name?
<qschulz>
if so, look at how it does it
<qschulz>
the adc-keys and gpio-keys driver seem to be setting input->name to that?
<qschulz>
sorry, gpio-keys seems to be doign that, not adc-keys
<qschulz>
mmm not sure this is reporting the button label though...
<qschulz>
only the name of the entire device
<naoki>
I don't know what the label property is for, but Documentation/devicetree/bindings/input/adc-keys.yaml does mention it. Is it strange that the adc-keys driver doesn't reference it?
<CounterPillow>
Right, I have a different patch to the adc-keys driver I should follow up on
<diederik>
mmind00: I guess it would be spacebar heating ... glad to guessed right :-D
ldevulder has quit [Ping timeout: 256 seconds]
<mmind00>
diederik: yep ... and judging by naoki 's additional comment about maskrom, this is a can of worms best left unopened ;-)
<diederik>
yeah
<mmind00>
because somewhere, some userspace might depend on the current setup ... and we can never know if and when that is the case
<mmind00>
and of course doing judgement-calls then just causes even more judgement-calls to be requested :-)
ldevulder has joined #linux-rockchip
* mmind00
can now go back to being sad that his rk3568 npu isn't working on first try :-)
<naoki>
haha
<CounterPillow>
mmind00: fairly different register layout than RK3588 as far as I've seen
<CounterPillow>
rk3576 has yet again a different layout, one that more closely aligns to rk3562 from what I can tell
<naoki>
Kwiboo: I can't get the E24C/E54C gmac and rtl8367rb to work
<naoki>
rk_gmac-dwmac ffbe0000.ethernet end0: Failed to reset the dma
<naoki>
rtl8365mb-mdio stmmac-0:1d: failed to get RESET GPIO
<naoki>
rtl8365mb-mdio stmmac-0:1d: probe with driver rtl8365mb-mdio failed with error -16
<mmind00>
CounterPillow: at least from comparing rk3568/rk3588 TRM register description it looked pretty similar?
* mmind00
may be blind though
<CounterPillow>
I didn't look too closely last time I poked at this, but they definitely have slightly different codepaths in the downstream kernel driver
<mmind00>
CounterPillow: yeah ... found that "scaling" for some value thing which seems to be the big difference ... but I'm stuck on a more basic level :-)
<CounterPillow>
My bet's on clocks
<mmind00>
probably
ldevulder has quit [Ping timeout: 256 seconds]
ldevulder has joined #linux-rockchip
<Kwiboo>
naoki: I will run some new tests on my E24C (and send an updated series) later today and/or tomorrow, do you have a tree/branch with your e54c DT?
<Kwiboo>
naoki: also ensure you have CONFIG_RESET_GPIO=n, else the rtl8365mb-mdio driver must be patched
<Kwiboo>
should complete and send my patches for that ;-)
<naoki>
Kwiboo: CONFIG_RESET_GPIO was set to "m" by default. When I tried to disable it with modprobe, I got the following error:
<naoki>
mdio_bus stmmac-0:1d: deferred probe pending: rtl8365mb-mdio: failed to get reset control
<naoki>
Kwiboo: rtl8365mb driver is now running with "net: dsa: realtek: fix reset-gpios handling", I can see wan and lan[1-3] interfaces, but I still get