17:26 <ukleinek> hi, I'm trying to get USB up and running on a helios64. A pluged usb device gets power, but lsusb only reports 2 root hubs.

17:26 <ukleinek> is there something obvious I might have missed?

17:27 <macc24> bruh

17:27 <ukleinek> for usb2 stuff I only need D+ and D-, VBUS and Ground, right? VBUS seems to be there.

17:29 <ukleinek> D+ and D- go to TYPEC1_DP and TYPEC_DN on the soc (rk3399)

20:14 <mmind00> ukleinek: in general, on non-chromeos devices the typec should simply always go to host-mode [missing integration into the kernel's typec-framework] ... so it should normally just work

20:15 <mmind00> at least it did in the past on other boards

20:34 <macc24> mmind00: do you know of any changes between 5.12-rc5 and 5.13-rc5 that could break mwifiex wifi card on pcie on rk3399 chromebook(kevin)?

20:35 <macc24> wifi broke for me on kevin, it looks like it fails around line 3095 in drivers/net/wireless/marvell/mwifiex/pcie.c

21:15 <ukleinek> mmind00: found my problem. I needed to use &usbdrd3_1, not the other one :-)

21:16 <mmind00> ukleinek: yay

21:17 <macc24> ,_,

21:17 <mmind00> macc24: I recall seeing patch concerning pcie

21:18 <mmind00> macc24: https://lore.kernel.org/linux-pci/20210607112856.3499682-1-punitagrawal@gmail.com/

21:20 <macc24> mmind00: ill try reverting it and ill report backl

21:26 <macc24> something didn't make it to mainline? https://github.com/torvalds/linux/blob/master/arch/arm64/boot/dts/rockchip/rk3399.dtsi#L230

21:34 <macc24> mmind00: found the issue

21:34 <macc24> this didn't make it to linux-git master branch https://lore.kernel.org/linux-pci/20210607112856.3499682-5-punitagrawal@gmail.com/

21:37 <macc24> brb

21:59 <macc24> yep that's what's missing

21:59 <macc24> now onto why it doesn't power off

22:09 <macc24> will investigate that tomorrow

22:37 <vagrantc> having mixed luck with nvme on pinebook pro ... seems to work fine as long as i don't initialize nvme from u-boot ... when nvme is initialized in u-boot, the nvme isn't recognized from the linux kernel

22:37 <vagrantc> but u-boot loads kernel/initrd/device-tree from nvme just fine ...

22:38 <vagrantc> 5.12.x kernel, fwiw

22:41 <vagrantc> and u-boot 2021.07-rc4

23:42 <punit> I wonder if you're seeing the same issue I reported many months back - http://u-boot.10912.n7.nabble.com/NVMe-boot-issues-on-RockPro64-td424863.html

23:43 <punit> tl;dr: uboot is able to access / load from NVMe but on exit it leaves the drive in a state Linux is not expecting

23:44 <vagrantc> that sounds very similar

23:45 <punit> From what I remember, it was the intersection of somewhat non-compliant SSD and uboot driver not handling interrupts properly

23:45 <punit> The problem went away when I switched the drive

23:47 <vagrantc> that doesn't bode well

23:50 <punit> By any chance do you have an ADATA SSD? Or it maybe the one using the same controller...

23:55 <vagrantc> cruicial

23:55 <vagrantc> er, crucial