chen_ has quit [Read error: Connection reset by peer]
lurchi__ has quit [Ping timeout: 248 seconds]
vstehle has joined #linux-rockchip
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-rockchip
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-rockchip
nighty- has joined #linux-rockchip
chewitt has joined #linux-rockchip
cnxsoft has quit [Quit: cnxsoft]
cnxsoft has joined #linux-rockchip
cnxsoft has quit [Remote host closed the connection]
BenG83 has joined #linux-rockchip
kbingham has joined #linux-rockchip
<kbingham>
Hi All - I was hoping to use an RK3399 to test UVC Video gagdet on a Type-C connector using the DWC3 ... can anyone confirm (or deny) that the type-c uses the DWC3 ? Any known issues there presently ?
cnxsoft has joined #linux-rockchip
<Ke>
at least on Kevin the type-c connector works
<kbingham>
On the RK3399 - is the FUSB just a driver for the Type-C conector ? or is it using drivers/usb/gadget/udc/fusb300_udc.c to implement the UDC ?
<kbingham>
If it's using fusb_udc ... that's the end of my adventure (for now )
<mmind00>
kbingham: the fusb does things like cable-state detection - similar to what the extcon_usbc_cros_ec does ... but so far nobody managed to make it work with the fusb
<kbingham>
Ok - so that could just be handling the Type-C connector ....
<mmind00>
kbingham: but one thing I thought of over coffee just now ... you could just make the rk3399-typec-phy driver default to device mode
<mmind00>
kbingham: i.e. in the patch I showed you in the mail, eballetbo added a default to make it work as host when the extcon is missing ... you could just hack the driver to make it default to device mode
<kbingham>
mmind00, Aha - sorry - only just saw your mail from 12:57 :D
<mmind00>
kbingham: no worries :-D
<kbingham>
if FUSB is just the PHY, and the UDC is DWC3 - then I can still continue :)
<mmind00>
kbingham: aka the phy has its own driver ... fusb does things like cable-state detection and negotiating the power-delivery
<mmind00>
(without the "aka" ;-) )
<kbingham>
mmind00, do you have console access into any rk3399 boards you could type "ls /sys/class/udc" perhaps ?
<mmind00>
kbingham: give me some minutes ... need to bring up the farm first
<kbingham>
feels like I'm still a way from booting this board. ... hrm ... I'm crazy - I can just boot the vendor kernel and test it myself.
<mmind00>
:-D
<kbingham>
pi@nanopi4:/sys/class/udc$ ls
<kbingham>
fe800000.dwc3
<kbingham>
Phew :)
<kbingham>
That removes my panic at least :D
<kbingham>
back to getting ML to boot.
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has quit [Remote host closed the connection]
<kbingham>
Aha - helps to get the speed right :) - the board uses console=ttyS2,1500000
<Kelsar>
kbingham: do you make writeup of the installtion on the nano pc? have one at home some weeks now, no time to get it running...
<kbingham>
Kelsar, Sure - will do.
<kbingham>
I've ditched the idea of porting the vendor kernel for now, and I'm now moving to the firefly dtb as a base... brings the kernel up to a point.
<kbingham>
Now I think i just need to map the MTD partitions the same as the vendor distro to get the filesystem back.
<Kelsar>
rk3399 is upstream... i build a kernel, just haven't tested it yet
<kbingham>
Kelsar, Sure - RK3399 is upstream - but the board support (DTB) for the nano-pc-t4 isn't.
<kbingham>
(Though I'll post patches when/if/as I get it working)
<Kelsar>
oh dear, i am new to all this ;)
<kbingham>
Kelsar, don't worry - you won't be in a year :D
<Kelsar>
kbingham: right now i battle other things, so i haven't much time at home. My Plan still is putting that thing into a cherry G80-3000
vagrantc has joined #linux-rockchip
<kbingham>
Argh ... the clock names for the DWC3 are wrong in the firefly DTB ...
<kbingham>
wait ... no - in the rk3399.dtsi ?
wadim_ has quit [Quit: Leaving.]
vagrantc has quit [Quit: leaving]
aalm has quit [Quit: xyz 2.0.1]
athidhep has quit [Ping timeout: 244 seconds]
athidhep has joined #linux-rockchip
afaerber has quit [Quit: Leaving]
afaerber has joined #linux-rockchip
aalm has joined #linux-rockchip
return0e has quit [Ping timeout: 256 seconds]
return0e_ has joined #linux-rockchip
<mmind00>
kbingham: not necessarily ... if dwc3 cries over a missing clock that is to be expected (ref clk I think) - and doesn't hinder it from working ... i.e. dwc3-of-simple seems to use different clock names or so
<kbingham>
mmind00, Yes, it does look like the clock names are abstracted and not important to match due to the dwc3-of-simple.
<kbingham>
I'm currently debugging the dwc3_core_get_phy() returning -EPROBE_DEFER constantly.
<kbingham>
Seems that maps through to the tcphy0_usb3 and tcphy1_usb3 nodes in rk3399.dtsi ... so I guess I haven't got those configured right yet :)
<mmind00>
or not compiled / probed ;-)
<mmind00>
i.e. make sure the optional-extcon patch is in the kernel you're using
<kbingham>
I'm on heiko/for-next (where heiko is your linux-rockchip.git)
<kbingham>
CONFIG_PHY_ROCKCHIP_TYPEC=y is definitely compiled in :)
<kbingham>
Ho hum. Times up for tonight ... I'll have to try to compare why the vendor kernel probes this but not ml :S
<mmind00>
kbingham: hmm, the extcon patch is old enough to be in the 4.18-rc1 that is based on .. strange
<kbingham>
Yes, I believe the patch is in ...
<kbingham>
But something isn't right :(
<kbingham>
But I was supposed to leave 13 minutes ago :)
<kbingham>
mmind00, Thankyou for your help today!
<mmind00>
kbingham: :-)
<kbingham>
I'll try the google extcon tomorrow. I've still been on fusb today :) - so perhaps the fault is somewhere in there.
<mmind00>
kbingham: you shouldn't need the extcon ... the type-c-phy should probe without it
<mmind00>
kbingham: ha ... make sure the usb2phy also gets probed
<mmind00>
kbingham: but then again, both phy drivers are part of the am64 defconfig, so they should at least get build
beeble has quit [Remote host closed the connection]
anarsoul|2 has joined #linux-rockchip
beeble has joined #linux-rockchip
xerpi has joined #linux-rockchip
afaerber has quit [Ping timeout: 276 seconds]
nighty- has quit [Quit: Disappears in a puff of smoke]
afaerber has joined #linux-rockchip
xerpi has quit [Remote host closed the connection]