mmind00 changed the topic of #linux-rockchip to: Rockchip development discussion | public log at https://libera.irclog.whitequark.org/linux-rockchip
kevery1 has joined #linux-rockchip
kevery has quit [Ping timeout: 244 seconds]
kevery1 is now known as kevery
stikonas has quit [Quit: Konversation terminated!]
<dsimic> naoki: is there a link to a ML discussion in which you proposed changes to the rock5a-rk3588s_defconfig and rock5b-rk3588_defconfig filenames in U-Boot?
kevery1 has joined #linux-rockchip
<dsimic> (you mentioned those changes as requested before, but refused)
kevery has quit [Ping timeout: 245 seconds]
kevery1 is now known as kevery
kevery1 has joined #linux-rockchip
kevery has quit [Read error: Connection reset by peer]
kevery1 is now known as kevery
<naoki> dsimic: I thought I discussed it on ML, but it seems wrong
<naoki> maybe I discussed it on here
<naoki> ah
<naoki> I don't understand why rock-5-itx is not rock5-itx
<naoki> (of course I prefer rock-5-itx)
<naoki> and I don't understand I have to use "rockpi-e-v3" instead of "rock-pi-e-v3" ;)
<naoki> I prefer -> Radxa's naming rule
<naoki> thare is no ROCK5A/ROCK5B
<dlg> linux developers know best
<naoki> best? is it better than fix request from maker?
<naoki> btw my E25 RGB LEDs light up (blue-ish) by default, I guess "pwm-leds-multicolor" doesn't turn off pwm
<naoki> "pwm-leds" doesn't have this issue
<naoki> does anyone have E25?
<dlg> i do, but not running linux
System_Error has quit [Remote host closed the connection]
System_Error has joined #linux-rockchip
<dsimic> naoki: why it isn't? because it is :)
<dsimic> IOW, there are no official rules, so it was simply named that way, just as the other two were named the other way
<dsimic> when it comes to the DT naming, I'd agree that "rockpi-e-v3" is a better choice, because it's consistent
hanetzer has joined #linux-rockchip
<dsimic> when it comes to the U-Boot defconfig naming, I'd prefer to see "rock-5b", because that would be more consistent
<dsimic> also, I think that saying "XYZ is better" isn't the best way for requesting some change
<dsimic> simply because "better" is subjective and rather vague
<dsimic> saying "it would be more consistent", for example, might be a better choicw
<dsimic> (I'm referring to your messages on the U-Boot ML)
<dsimic> s/choicw/choice/
warpme has joined #linux-rockchip
<diederik> naoki: if you're going to make a v2 for your multicolor patch set ... maybe also explain *why* you want to add multicolor to the defconfig?
<diederik> haha, looks like Krzysztof had the same thought :)
eballetbo has joined #linux-rockchip
UndrWater has quit [Quit: ZNC 1.8.2+deb2build5 - https://znc.in]
UndrWater has joined #linux-rockchip
<naoki> diederik: :)
<naoki> dsimic: more consistent, noted ;)
<naoki> anyway it seems I need to check pwm-leds-multicolor init function(?) first
<naoki> no configuration, but it lights up, like a ghost ;)
raster has joined #linux-rockchip
<warpme> guys: what will be easiest way to have led trigger working on emmc AND sd card in the same time?
<Kwiboo> naoki: regarding rock-pi-s and wifi, at least last time I tested it seemed to be working, yesterday my kernel crashed for other reasons, but I can see "rtw_8723ds mmc2:0001:1: Firmware version 48.0.0, H2C version 0" and "Bluetooth: hci0: RTL: fw version 0xaa5d675f"
<Kwiboo> the kernel and DT change to make it work/be detected was to ensure io-domain voltage is correct, with both io-domain and DT changes I would expect it to work
<Kwiboo> using rtw88/rtw8723d_fw.bin from linux-firmware and rtl_bt/rtl8723ds_{config,fw}.bin from https://github.com/radxa-pkg/radxa-firmware/tree/main/firmware/rtl_bt
<naoki> Kwiboo: for OpenWrt, I was cherry-picked Linux changes, but I didn't it U-boot side, I'm not sure both should be patched
<Kwiboo> naoki: if io-domain driver+DT is picked for u-boot it does not have to also be configured by kernel, the wifi DT change must be included in the DT used by kernel
<Kwiboo> if I understand correctly OpenWrt use a FIT with kernel + DT, if that is the case the U-Boot DT should not matter in kernel
<mmind00> warpme: use the "disk-activity" trigger?
<warpme> mmind00 : i'm a bit lost with this: looking in /sys/class/leds/led1/trigger i see following possible triggers: none kbd-scrolllock kbd-numlock kbd-capslock kbd-kanalock kbd-shiftlock kbd-altgrlock kbd-ctrllock kbd-altlock kbd-shiftllock kbd-shiftrlock kbd-ctrlllock kbd-c
<warpme> trlrlock heartbeat cpu cpu0 cpu1 cpu2 cpu3 cpu4 cpu5 cpu6 cpu7 activity default-on netdev pattern [mmc1] mmc0 rc-feedback rfkill-any rfkill-none bluetooth-power
<warpme> hci0-power rfkill0 rfkill1 stmmac-0:01:link stmmac-0:01:1Gbps stmmac-0:01:100Mbps stmmac-0:01:10Mbps
<warpme> and there is no "disk-activity".....
<mmind00> you want CONFIG_LEDS_TRIGGER_DISK enabled in your kernel config
<naoki> Kwiboo: thanks, I think my work should be okay (but it's not working lol)
<naoki> it needs more investigation
<naoki> hmm... we need to maintain local patches to use mainline things for our purpose...
<naoki> hmm, referring vendor kernel is not permitted...?
<diederik> That 'some random kernel on the internet' has something is not a reason itself for a change
<diederik> The underlaying reason why that 'random' kernel has that, could be a proper reason
<naoki> I'm not talking about 'some random kernel on the internet'...
<naoki> I think I saw several patches like "vendor kernel does XXX, so we do it too"
<diederik> I deliberately put 'random' in quotes because I know it's not completely/actually random
<diederik> And the patches you're likely referring to where voltage changes done my Rockchip themselves and then the 'assumption' was that if anyone knows what the best/proper voltages are, it's Rockchip
<diederik> s/my/by/
<naoki> anyway "we(Radxa) need it" is not proper reason
<naoki> we cannot change behavior for our products
Stat_headcrabed has joined #linux-rockchip
<warpme> mmind00: this is what i suspect (missing in config). added in .config; rebuild and can't get it working. Isn't disk-activity works only for ata? (i see kernel has dependency on CONFIG_ATA)
<warpme> also i tried with adding "led-triggers = <&sdmmc>, <&sdhci>;" nothing on led :-(
<mmind00> warpme: strange ... I must have things confused :-( ... because part of me wanted to remember that also working for mmc ... but I must have remembered wrong
<mmind00> interestingly ata has the opposite problem ... there is no per ata-host led support so far
<warpme> :-)
<mmind00> warpme: you could try to add calls to ledtrig_disk_activity for mmc ;-)
<mmind00> i.e. it does make sense to also allow a generic disk-activity led like cases have to tackle mmc hosts too
<warpme> exactly! My use-case is: user flashes OS on sd card (and have led showing access to sdcard); then user installs OS in emmc; the removes sd card; boots OS form emmc. all this works well for me except OS working from sd card has no disk activity led.....
<warpme> s/from sd card has no disk activity led...../from emmc has no disk activity led...../g
<mmind00> warpme: so I guess just add a led-call for the disk-activity into https://elixir.bootlin.com/linux/v6.10.5/source/drivers/mmc/core/core.c#L177 ... similar to how its done in ata https://elixir.bootlin.com/linux/v6.11-rc3/source/drivers/ata/libata-core.c#L4850 ... in theory mmc maintainers should hopefully pick such a patch
<warpme> let me try :-)
Stat_headcrabed has quit [Quit: Stat_headcrabed]
<diederik> mmind00: finally! ;-P
dsimic has quit [Ping timeout: 245 seconds]
<mmind00> diederik: finally what?
dsimic has joined #linux-rockchip
<diederik> merging the wolfvision pf5 visualizer display (or do I remember that incorrectly?)
<mmind00> diederik: ah that, yeah :-)
<mmind00> needing a dtc change was "fun" :-D
<diederik> indeed ;)
psydroid has joined #linux-rockchip
Perflosopher has quit [Read error: Connection reset by peer]
Perflosopher0 has joined #linux-rockchip
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
psydroid has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/]
valpackett has quit [Remote host closed the connection]
valpackett has joined #linux-rockchip
psydroid has joined #linux-rockchip
psydroid has quit [Excess Flood]
psydroid has joined #linux-rockchip
crabbedhaloablut has quit []
crabbedhaloablut has joined #linux-rockchip
raster has quit [Quit: Gettin' stinky!]
Stat_headcrabed has joined #linux-rockchip
stikonas has joined #linux-rockchip
Stat_headcrabed has quit [Quit: Stat_headcrabed]
linkmauve has quit [Ping timeout: 248 seconds]
vagrantc has joined #linux-rockchip
Stat_headcrabed has joined #linux-rockchip
ungeskriptet has quit [Ping timeout: 255 seconds]
ungeskriptet has joined #linux-rockchip
System_Error has quit [Ping timeout: 260 seconds]
ungeskriptet7 has joined #linux-rockchip
ungeskriptet has quit [Ping timeout: 260 seconds]
ungeskriptet7 is now known as ungeskriptet
Stat_headcrabed has quit [Quit: Stat_headcrabed]
ungeskriptet has quit [Ping timeout: 252 seconds]
System_Error has joined #linux-rockchip
eballetbo has quit [Quit: Connection closed for inactivity]
ungeskriptet has joined #linux-rockchip
System_Error has quit [Remote host closed the connection]
System_Error has joined #linux-rockchip
kevery has quit [Ping timeout: 252 seconds]
kevery has joined #linux-rockchip
<naoki> personally, if there is free PWM pin, LED should be connected to it instead of GPIO only (non-PWM) pin
<naoki> some LEDs are toooooooooooo bright for my eyes...
<naoki> (let's check schematics...)
<naoki> Kwiboo: ROCK S0 PWR_LED is PWM LED! ;)
<naoki> (I'm not talking about user configuration)
shailangsa has joined #linux-rockchip
<naoki> I guess ROCK 3A/3B/5A/5B heartbeat LED can be PWM LED... heartbeat can be weak!
hanetzer has quit [Quit: WeeChat 4.3.4]
System_Error has quit [Remote host closed the connection]
System_Error has joined #linux-rockchip