narmstrong changed the topic of #linux-amlogic to: Amlogic mainline kernel development discussion - our wiki http://linux-meson.com/ - ml linux-amlogic@lists.infradead.org - official channel moved from Freenode - publicly logged on https://libera.irclog.whitequark.org/linux-amlogic
naoki has joined #linux-amlogic
buzzmarshall has joined #linux-amlogic
psydroid has quit [Read error: Connection reset by peer]
vagrantc has quit [Ping timeout: 265 seconds]
hexdump0815 has quit [Ping timeout: 244 seconds]
hexdump0815 has joined #linux-amlogic
psydroid has joined #linux-amlogic
buzzmarshall has quit [Quit: Konversation terminated!]
jacobk has quit [Ping timeout: 276 seconds]
jacobk has joined #linux-amlogic
naoki has quit [Quit: naoki]
runxiyu has joined #linux-amlogic
naoki has joined #linux-amlogic
naoki has quit [Client Quit]
jacobk has quit [Ping timeout: 260 seconds]
jacobk has joined #linux-amlogic
vagrantc has joined #linux-amlogic
vagrantc has quit [Ping timeout: 252 seconds]
zsoltiv_ has joined #linux-amlogic
djrscally has joined #linux-amlogic
<f_> btw, just to make sure: gxl dwc can do UDC in mainline, right?
<f_> Somehow I can't really get it working
<f_> there's nothing in /sys/class/udc/, while there was something before there when I booted with the vendor u-boot from a USB stick (and it was why my dmesg kept getting spammed with "new device is full-speed" :p)
<f_> Because I don't have Wi-Fi working currently, and also because a USB hub is no longer an option I'd like to do usb internet, that's why I'm asking
<f_> ums/dfu and others work fine in mainline u-boot, BTW
<xdarklight> f_: Meson8b and GXBB can do it - so I'd be surprised if it doesn't work on GXL
<xdarklight> f_: actually I think I've tested it with Neil before around upstream commit a9fc15e0fd78ac57535c2e4fcddfafd9df32e37a ("usb: dwc3: meson-g12a: add support for GXL and GXM SoCs")
<f_> xdarklight: maybe I'm doing something wrong then?
<f_> hm
<f_> [ 276.554064] dwc3-meson-g12a d0078080.usb: Broken manual OTG switch
<f_> [ 276.554220] dwc3-meson-g12a d0078080.usb: switching to Device Mode
<xdarklight> the "broken OTG switch" message it's normal. it's a quirk of the SoC
<xdarklight> do you have CONFIG_USB_DWC3_DUAL_ROLE=y, =m or =n?
<f_> It's =y
vagrantc has joined #linux-amlogic
<xdarklight> sounds fine so far. can you share your dmesg?
<f_> Sure, moment.
<xdarklight> also do you have USB_DWC2_DUAL_ROLE=y, =m or y?
<f_> Also =y
<xdarklight> good. the peripheral part is managed by dwc2 (dwc3 is in host-only mode)
<f_> had trouble with paste.debian.net :p
<f_> the Bluetooth stuff you can ignore
<f_> Lines prefixed with `[pmOS-rd]` are the initramfs logs
<xdarklight> dwc2 c9100000.usb: dwc2_core_reset: HANG! Soft Reset timeout GRSTCTL_CSFTRST
<f_> Aha
<f_> I completely missed that bit
<f_> I might have an idea why it doesn't like the USB but uh, highly doubt
<f_> apparently the usbdl vuln exploitation leaves the USB in a weird state ... but don't think that'd make sense that linux would complain
<f_> AFAIK it re-inits USB, right? And it works fine in u-boot anyway
<xdarklight> GXL is a bit weird, there's only a shared reset line (see arch/arm64/boot/dts/amlogic/meson-gxl.dtsi near the top) for USB
<f_> mmhmmm
<xdarklight> https://github.com/endlessm/linux-s905x/blob/master/arch/arm64/boot/dts/amlogic/gxl_skt.dts#L847-L852 - I think we don't specify all of them in mainline. However, it's a bit odd since https://github.com/endlessm/linux-s905x/blob/master/include/dt-bindings/reset/aml_gxl.h and https://github.com/endlessm/linux-s905x/blob/master/include/dt-bindings/reset/aml_gxbb.h are overall identical but the offsets are way different from what we have upstream (
<xdarklight> which may just be a different way of counting, but I'm not sure)
* f_ looks at aquaman downstream dt dump
<xdarklight> ah wait, the numbers are likely to be different because those are gate clocks, not resets
<f_> yep they're `clocks` in the dump I have :p
<xdarklight> if they're clocks: compare /sys/kernel/debug/clk/clk_summary on working and non-working boots
<xdarklight> and how stupid of me to think that resets would be resets. of course they're clocks!
<f_> well sadly I was stupid enough to wipe the eMMC entirely including the vendor u-boot and android
<f_> I have backups, but when I checked the vendor u-boot wasn't quite happy with pmOS being on the eMMC (it completely crashed when I tried booting it up)
<xdarklight> I wouldn't call that stupid, just makes things harder. do you have another GXL device at hand where you can compare "grep usb /sys/kernel/debug/clk/clk_summary" with the problematic box?
<f_> I do, thankfully
<f_> > I wouldn't call that stupid, just makes things harder.
<f_> It is stupid also because the stick has amlogic secureboot enabled
<f_> xdarklight: will do later. For now I'm going afk
<f_> \o
<f_> and thanks for the help so far
<xdarklight> good luck :-) I'll be heading to bed soon
naoki has joined #linux-amlogic
psydroid has quit [Quit: KVIrc 5.2.6 Quasar http://www.kvirc.net/]
djrscally has quit [Ping timeout: 248 seconds]