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 - Publicly Logged on https://irclog.whitequark.org/linux-amlogic
TheAssassin has joined #linux-amlogic
vagrantc has quit [Quit: leaving]
vicencb has quit [Quit: Leaving.]
<kgfh> Good to know, I'll see if I can come up with something this weekend.
BlueMatt has quit [Ping timeout: 240 seconds]
khilman has quit [Ping timeout: 240 seconds]
khilman has joined #linux-amlogic
BlueMatt has joined #linux-amlogic
Linnaea_ has quit [Ping timeout: 240 seconds]
Linnaea_ has joined #linux-amlogic
buzzmarshall has quit [Remote host closed the connection]
cmeerw has joined #linux-amlogic
kaspter has joined #linux-amlogic
damex_ has joined #linux-amlogic
damex has quit [Ping timeout: 256 seconds]
ldevulder has quit [Ping timeout: 264 seconds]
yann has joined #linux-amlogic
sputnik_ has quit [Ping timeout: 258 seconds]
trem has joined #linux-amlogic
buzzmarshall has joined #linux-amlogic
lioka has quit [Quit: leaving]
armoon has joined #linux-amlogic
vicencb has joined #linux-amlogic
Lyude has quit [Quit: WeeChat 2.8]
Lyude has joined #linux-amlogic
ar-rahman has joined #linux-amlogic
<ar-rahman> hi guys
<ar-rahman> how are yall
<ar-rahman> I was wondering if the USB bug in Odroid C1 with mainline was fixed?
<armoon> USB on C1+ and C2, I thinks it's not fix yet completely.
<ar-rahman> wow
<ar-rahman> this has been hanging around for many years
<ar-rahman> How come that it was working with 3.10
<ar-rahman> and just stopped working with mainline?
<ar-rahman> what makes it just brake ?
<ar-rahman> doing docker stuffs on microsd card will kill it
<ar-rahman> that's why external HD are vital
<armoon> I feel their will be fix some time soon but mean while wout can check https://github.com/xdarklight/linux latest branch he has some more patches which are not upstream yet
armoon has quit [Remote host closed the connection]
<ar-rahman> Ok xdarklight has made some progress you mean ?
<kgfh> Is there any progress on enabling the discard capability of the mmc controller?
<kgfh> I found some mentions of this by searching for irc logs but so far it still doesn't seem to be enabled.
armoon has joined #linux-amlogic
armoon has quit [Remote host closed the connection]
vagrantc has joined #linux-amlogic
<xdarklight> armoon: ar-rahman: I'm currently not actively working on finding the root cause of this USB issue. it only seems to happen when that on-board USB hub is involved (which not all of my boards have) so I'm thinking that it probably has something to do with the dwc2 driver. so I recommend that someone gets in touch with the dwc2 maintainers to analyze and solve this
<xdarklight> kgfh: I think for 5.8 or 5.9 (I don't remember) discard support is enabled for all MMC controller drivers
<kgfh> Nice!
<kgfh> This is honestly the very last road block for me to consider the glx platform to be "fully functional" for my use cases.
<kgfh> *gxl
<kgfh> Lately I've also been tinkering with the fedora aarch64 image, and they seem to be using the u-boot -> shim -> grubaa64 -> kernel approach, and I'm having a hard time to get it to boot on a gxl board with mainline u-boot.
<ar-rahman> xdarklight: The problem seems to be on booting with an external HD attached
<ar-rahman> The board doesn't recognize it
<ar-rahman> I tried some version of armbian
<ar-rahman> and I could make the HD recognized whén connected with another usb stick together
<ar-rahman> Also I had lots of problem with the LAN adapter
<ar-rahman> I could from time to time access the board via SSH
<ar-rahman> but not all the time
<xdarklight> ar-rahman: wait. only one problem at a time please ;-)
<xdarklight> ar-rahman: I don't know what Amlogic used in their 3.10 kernel to make USB work. lots of code in their 3.10 kernel drivers is not using upstream frameworks and so much code is duplicated by using custom drivers
<ar-rahman> How about connecting with those dudes ?
<xdarklight> ar-rahman: debugging will require knowledge about the inner workings of the dwc2 (USB) controller - which I don't have. that's why I suggest to contact the dwc2 maintainers
<ar-rahman> Maybe it was so simple they could tell
<ar-rahman> I don't know the dwc2 meaning
<xdarklight> ar-rahman: last time I heard from someone at Amlogic was more than 6 months ago. since then there's silence - so I have nobody to ask currently
<xdarklight> ar-rahman: dwc2 is the name of the USB controller used on the S805 and S905 (as well as many other) SoCs
<xdarklight> (the actual name is much longer, if you're interested: https://www.synopsys.com/dw/ipdir.php?ds=dwc_usb_2_0_hs_otg )
<ar-rahman> Oh thanks for clarification :D
<ar-rahman> Did you talk with the guys in this repo ? https://github.com/torvalds/linux/tree/master/drivers/usb/dwc2
<xdarklight> ar-rahman: that's what I meant with "dwc2 maintainers" :-) the following fix was made two years ago: https://www.spinics.net/lists/linux-usb/msg174990.html - I'm pretty sure that USB devices were detected with this patch back then. the relevant code-change is still in the latest mainline kernel but for some reason it's not working anymore. unfortunately I'm currently very limited in terms of spare-time, so that's why I won't contact
<xdarklight> them now
<ar-rahman> how about I post an issue on their ticket list with the link to this thread
<xdarklight> ar-rahman: I'
<xdarklight> I'm not sure if there's any ticket system
<ar-rahman> Lol that's not the case
<ar-rahman> Rofl
<xdarklight> ar-rahman: the way that I'm aware of is a discussion on the mailing list: https://pastebin.com/a5VWVbUZ
<xdarklight> Minas is the maintainer, the rest are "interested parties" who should be Cc'ed
Darkmatter66_ has quit [Ping timeout: 240 seconds]
<xdarklight> ar-rahman: regarding the Ethernet problem you mentioned: this has probably been fixed with Linux 5.8 and this patch (this cannot be applied by itself to 5.7, there are some other patches as dependencies): https://github.com/torvalds/linux/commit/005231128e9e97461e81fa32421957a7664317ca
<ar-rahman> I would say you are kind of a ninja finding the clues for everything men
<ar-rahman> That's an impressive skill
<ar-rahman> Thumbs up
<ar-rahman> You are well aware of the internals of the ARCANE :D
Darkmatter66 has joined #linux-amlogic
<xdarklight> :-) it's four years now that I'm working on mainline support for Amlogic SoCs together with general interest in why and how hardware works made me discover some interesting things over time
<ar-rahman> That's enoough time to develop ninja skills for sure
<ar-rahman> For my part I got lately the UART usb adapter to listen in the debug terminal of the boards
<ar-rahman> But I don't know how it works or how to make it work
<ar-rahman> I didn't know where to find the infos to setup something
<kgfh> Looking at this tv box I have, it seems to use a GPIO to control the 5V power supply (USB) and the 3v3 on HDMI. What would be the best approach to put this into my dts meant for mainline?
<kgfh> The recovery button behind the CVBS port is also attached to a GPIO.
<kgfh> I've figured out the IR receiver's GPIO as well, is this standard, or do I actually need to define this in my dts?
<kgfh> (IR is GPIOAO_7)
<ar-rahman> xdarklight: It is good news for the LAN adapter bug fix
sputnik_ has joined #linux-amlogic
<kgfh> I also have two GPIO R/G LED clusters, they can't be turned off, I can only flip the colour by toggling the GPIOs. How can I put them to good use in the device tree, if it's possible? One cluster is for power and the other one is for WiFi.
warpme_ has quit [Quit: Connection closed for inactivity]
<xdarklight> kgfh: regarding the GPIO for 5V: my general suggestion is to look at other board.dts to see if there's something similar (for example arch/arm64/boot/dts/amlogic/meson-gxl-s905x-p212.dtsi has a HDMI_5V regulator)
<xdarklight> kgfh: the IR receiver has to be enabled. the SoC has a hardware decoder for IR, so not sure if you need to go through some GPIO polling or whether the hardware one can be used
<xdarklight> kgfh: multi-color LEDs are a very new thing to the Linux kernel. I think 5.9 finally gained support for some of that, but I think the GPIO based ones don't allow multi-colors yet
_whitelogger has joined #linux-amlogic
kgfh has quit [Quit: Connection closed for inactivity]
Tonymac32 has quit [Read error: Connection reset by peer]
cmeerw has quit [Ping timeout: 246 seconds]
yann has quit [Ping timeout: 240 seconds]