ChanServ changed the topic of #linux-rockchip to: Rockchip development discussion | IRC log http://irclog.whitequark.org/linux-rockchip | Community GH https://github.com/linux-rockchip | Rockchip GH https://github.com/rockchip-linux | ML https://groups.google.com/group/linux-rockchip
<anarsoul> ah, looks like I need couple es8316 patches
<anarsoul> nope, already have them :\
kevery has joined #linux-rockchip
kevery1 has joined #linux-rockchip
kevery has quit [Ping timeout: 240 seconds]
kevery1 is now known as kevery
lkcl has quit [Ping timeout: 276 seconds]
field^Mop has quit [Ping timeout: 240 seconds]
kaspter has joined #linux-rockchip
vstehle has quit [Ping timeout: 276 seconds]
stikonas has quit [Remote host closed the connection]
vagrantc has quit [Quit: leaving]
return0e has quit [Ping timeout: 240 seconds]
return0e has joined #linux-rockchip
return0e has quit [Read error: Connection reset by peer]
return0e has joined #linux-rockchip
levd has quit [Quit: Goodbye!]
levd has joined #linux-rockchip
return0e has quit [Ping timeout: 276 seconds]
return0e has joined #linux-rockchip
BenG83 has quit [Remote host closed the connection]
camus has joined #linux-rockchip
kaspter has quit [Ping timeout: 276 seconds]
camus is now known as kaspter
vstehle has joined #linux-rockchip
chewitt has joined #linux-rockchip
camus has joined #linux-rockchip
kaspter has quit [Ping timeout: 240 seconds]
camus is now known as kaspter
_whitelogger has joined #linux-rockchip
<mmind00> anarsoul: very strange ... I just did some reboots on rk3288, rk3328, rk3399 and px30 and didn't see any iommu hickups ... so I guess first step might be to find out what is actually different on your board that you're triggering that issue and how to reproduce it
levd has quit [Ping timeout: 265 seconds]
return0e_ has joined #linux-rockchip
jelly has quit [Read error: Connection reset by peer]
return0e has quit [Ping timeout: 240 seconds]
levd has joined #linux-rockchip
jelly-home has joined #linux-rockchip
return0e has joined #linux-rockchip
jelly-home is now known as jelly
return0e_ has quit [Ping timeout: 265 seconds]
jelly has quit [Ping timeout: 240 seconds]
camus has joined #linux-rockchip
<anarsoul> mmind00: that's just a regular rockpro64 board
<anarsoul> also I'm using mainline u-boot
kaspter has quit [Ping timeout: 246 seconds]
camus is now known as kaspter
<mmind00> anarsoul: that's what I also guessed ... which makes it so puzzling
<anarsoul> oh
<anarsoul> also I'm using uart @ 38400 baud
<anarsoul> since I don't have one that supports 1500000 :)
<anarsoul> that may slow down things enough to reproduce the issue
<anarsoul> (I wonder if uart is using dma which may use iommu?)
_kj_ has quit [Ping timeout: 244 seconds]
jelly-home has joined #linux-rockchip
<mmind00> anarsoul: Rockchip SoCs have per-device-iommus ... aka each vop, the camera interface etc, each brings its own iommu instance ... all boards I tested except the rk3328 also run only at 115200
<mmind00> anarsoul: but what you could try is inserting a dev_info into the rk_iommu_shutdown function, so that we may see the iommu instance creating the issue
<mmind00> anarsoul: i.e. right now I'm always thinking it must be the VOP's iommu, but it can also be a different one I guess
jelly-home is now known as jelly
<micken> what is LUT ram in vop. Linux enables it (dumped sys_ctrl)
adjtm has quit [Ping timeout: 252 seconds]
ldevulder__ has joined #linux-rockchip
ldevulder_ has quit [Ping timeout: 265 seconds]
_whitelogger has joined #linux-rockchip
lopsided98_ has joined #linux-rockchip
lopsided98 has quit [Ping timeout: 245 seconds]
adjtm has joined #linux-rockchip
matthias_bgg has joined #linux-rockchip
kaspter has quit [Quit: kaspter]
kaspter has joined #linux-rockchip
vicencb has joined #linux-rockchip
return0__ has joined #linux-rockchip
return0e has quit [Ping timeout: 268 seconds]
vicencb has quit [Ping timeout: 276 seconds]
vicencb has joined #linux-rockchip
wadim_ has joined #linux-rockchip
matthias_bgg has quit [Quit: Leaving]
<DuClare> Mmm, I wonder if there's any rationale for the choice of defaulting to gpio otp in tsadc?
<DuClare> I tried cru on rk3288 and it seems to do the job just fine
field^Mop has joined #linux-rockchip
warpme_ has joined #linux-rockchip
vicencb has quit [Quit: Leaving.]
chewitt has quit [Ping timeout: 240 seconds]
ldevulder__ is now known as ldevulder
kaspter has quit [Remote host closed the connection]
kaspter has joined #linux-rockchip
matthias_bgg has joined #linux-rockchip
Substring has joined #linux-rockchip
vicencb has joined #linux-rockchip
wadim_ has quit [Quit: Leaving.]
JohnDoe_71Rus has joined #linux-rockchip
vicencb has quit [Quit: Leaving.]
vicencb has joined #linux-rockchip
drrty has joined #linux-rockchip
lkcl has joined #linux-rockchip
<robmur01> anarsoul: do you by any chance have CONFIG_DEBUG_SHIRQ enabled?
BenG83 has joined #linux-rockchip
<robmur01> that would seem to explain why the handler gets called while freeing the irq
lkcl has quit [Ping timeout: 240 seconds]
Substring has quit [Quit: Leaving]
<robmur01> hmm, in fact I think the WARN_ON(pm_runtime_get_if_in_use() == 0) case has always been nonsense
lkcl has joined #linux-rockchip
inode has quit [Ping timeout: 245 seconds]
return0e has joined #linux-rockchip
return0__ has quit [Ping timeout: 240 seconds]
<anarsoul> robmur01: yeah, it's enabled
<anarsoul> I'll disable it
<robmur01> cool, thanks for confirming
<robmur01> I'll spin a patch to shut it up - want a reported-by?
<anarsoul> yeah, why not
<anarsoul> I wonder if changing audio-supply domain to 3.3v was a mistake
<anarsoul> my rockpro64 can't talk to audiocodec :\
<robmur01> sounds familiar - IIRC there was something similar on rock64 where part of the schematic claimed the wrong voltage too
<anarsoul> it likely reminds you 798689e45190756c2eca6656ee4c624370a5012a
vagrantc has joined #linux-rockchip
<robmur01> oh, indeed it was rockpro64, ha!
<anarsoul> but it doesn't work for me even with this patch
* robmur01 wonders if they might have done a sneaky rev of the board
<robmur01> hmm, or come to think of it that *is* our old friend VCC_3V0 again...
<anarsoul> ?
<robmur01> pretty sure that was the one involved in the screwed-up GPIO voltage, wasn't it?
<anarsoul> that was gpio1830-supply
<anarsoul> it's set correctly in dts
inode has joined #linux-rockchip
stikonas has joined #linux-rockchip
<anarsoul> my
<anarsoul> my es8316 seems to be sitting on address 0x10, not 0x11
<anarsoul> :\
nsaenz has quit [Remote host closed the connection]
nsaenz has joined #linux-rockchip
<robmur01> heh, the datasheet does say that the LSB of the address is AD0, but then leaves me with no idea what AD0 means... :/
<anarsoul> yeah, changed it in dts and it seems to work
<anarsoul> that's weird
<anarsoul> oh
<anarsoul> see schematics
<anarsoul> CE pin that selects i2c address goes to gnd
<robmur01> ah, there it is indeed. Maybe your board somehow ended up with R225 populated instead of R226?
nsaenz has quit [Read error: Connection reset by peer]
<anarsoul> well
<anarsoul> r226 seems to be correct and it selects address 0x10, doesn't it?
<anarsoul> but we have 0x11 in dts
nsaenz has joined #linux-rockchip
<robmur01> s/your board/someone else's board/... reading is hard :P
ldevulder has quit [Quit: Leaving]
robmur01_ has joined #linux-rockchip
robmur01 has quit [Read error: Connection reset by peer]
BenG83 has quit [Quit: Leaving]
matthias_bgg has quit [Ping timeout: 240 seconds]
JohnDoe_71Rus has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/]
<anarsoul> reboot and poweroff don't work properly because I'm booting in efi mode :\
<anarsoul> looks like u-boot implements efi_reset_system via psci
<anarsoul> so likely it's broken in mainline ATF?
adjtm_ has joined #linux-rockchip
adjtm has quit [Ping timeout: 240 seconds]
<anarsoul> mmind00: any ideas where it should be fixed?
<mmind00> anarsoul: not sure ... at least from the kernel reset via psci seems to work
<anarsoul> reset or poweroff? :)
<mmind00> reset
<anarsoul> are you booting in efi mode?
<mmind00> nope
<anarsoul> mainline u-boot (with SPL) and ATF?
<mmind00> most boards, yeah
<mmind00> also the rk3399 chromeos devices
<anarsoul> guess it's easier for me to fix it in u-boot
<anarsoul> ATF doesn't have rk808 driver
<anarsoul> "rockchip_soc_system_off" is pretty much while(1) if you don't have a gpio to turn off power
<anarsoul> soc_global_soft_reset() should work though, but it's extremely unreliable for me
<mmind00> ATF also doesn't have an i2c-driver to begin with ... I think in the kernel a system-specific poweroff trumps the psci method
<anarsoul> mmind00: efi inits earlier and sets pm_power_off, rk808 then checks that pm_power_off is set and doesn't override it
nsaenz has quit [Ping timeout: 240 seconds]
<anarsoul> so unfortunately system-specific poweroff doesn't trump efi (which in its turn calls psci)
<mmind00> :-(
<anarsoul> moreover it doesn't even trump psci without efi
<anarsoul> so poweroff won't work on mainline kernel with mainline u-boot with mainline ATF
<mmind00> anarsoul: ah, I think I remember Chromebooks using gpio-poweroff ... which always sets pm_power_off
<mmind00> it seems
<anarsoul> and it'll work even with psci
<anarsoul> since it's handled in ATF
<mmind00> ATF does gpio-poweroff?
<anarsoul> yes :)
<anarsoul> and reset
<mmind00> so I guess "someone" could do an i2c driver port and a minimal rk808 in ATF that at least handles the poweroff
<anarsoul> mmind00: easier to handle that in u-boot
<anarsoul> at least for efi case
<anarsoul> since it goes linux -> efi (which is u-boot) -> psci
<anarsoul> u-boot has rk808 and i2c drivers
<mmind00> spoilsport :-D
<anarsoul> moreover atf is generic for all the boards
<anarsoul> so I'm not sure how it's supposed to handle different power_off methods
<mmind00> ATF can handle devicetrees ... though I'm not sure if it has some sort of driver model
<mmind00> that was fun seeing in optee, that it actually creates device-instances from the dt and has an actual driver model
<anarsoul> I'll fix it in u-boot for now
<anarsoul> whoever boots without efi would have to fix it in ATF
<mmind00> :-D
<anarsoul> oh, i2c driver doesn't have master_xfer_atomic :\
<anarsoul> how is it supposed to work for poweroff then?
<anarsoul> :)
vicencb has quit [Quit: Leaving.]
BenG83 has joined #linux-rockchip
BenG83 has quit [Quit: Leaving]
warpme_ has quit [Quit: Connection closed for inactivity]