<phh>
mmind00: you mentioned testing suspend on rk3288. is suspend supposed to work with current mainline on rk3288 on some device?
<mmind00>
phh: yes ... (light) suspend was working at some point at least on veyron devices, last I heared there were issues but I haven't had time to investigate yet
<phh>
what's light suspend? clock off but not power?
* vagrantc
never managed to get the veyron-speedy to resume from suspend
<phh>
my speedy has his black going white... and that's all, never goes back
<mmind00>
phh: somewhat like that ... i.e. DDR still running etc ... deeper suspend needs the PIE code that is currently being worked on - for the code that needs to live in sram to restart the ddr
<phh>
I see
<mmind00>
and as I said, it was working at some point in the past, but somehow cannot resume right now
<phh>
doing echo mem > /sys/power/state switches to whichever mode is supported by the running kernel?
<mmind00>
phh: yep
<phh>
ok thanks, I'll see if a serial says anything
<mmind00>
phh: you probably already know that, but don't forget "no_console_suspend" for that
<phh>
erf, even though I do know that option I would have indeed forgot it, thanks
<phh>
got a null pointer derefencement in brcmf_ops_sdio_resume
nighty has quit [Quit: Disappears in a puff of smoke]
nashpa has quit [Ping timeout: 268 seconds]
nashpa has joined #linux-rockchip
<phh>
ok, brcmf_ops_sdio_resume crash is just because driver had no firmware loaded, so it was in a weird state... so everything works properly with serial debug.
<phh>
and without serial debug, it looks like the EC doesn't even try to wake up the cpu (led keeps blinking)
<phh>
right, that happens only when I force usb0 to device mode
<phh>
so... actually... suspend is working properly on veyron.
<mmind00>
phh: suspend working sounds nice :-)
<phh>
still got some issues, but I'll try to test properly, multiple times, to reduce my spam here
<phh>
ok, so wifi won't get back up (every request times out), even without keep-power-in-suspend and unbinding/rebinding. and it looks like that after a reboot instead of a fresh start, ec is in a weird state. i'll try to debug that more tomorrow