+- PINE64 (https://forum.pine64.org)
+-- Forum: Pinebook Pro (https://forum.pine64.org/forumdisplay.php?fid=111)
+--- Forum: Linux on Pinebook Pro (https://forum.pine64.org/forumdisplay.php?fid=114)
+--- Thread: A true mainline Linux Kernel for the Pinebook Pro (/showthread.php?tid=8207)
RE: A true mainline Linux Kernel for the Pinebook Pro - tsys - 11-08-2019
(11-08-2019, 04:39 AM)danielt Wrote: I'll go looking for more clues later today but I just wondered if you have seen anything like this yourself as you've been working.
Yeah, interestingly I had one similar event last evening. Upon starting the X user session from lightdm the eDP link failed and was unable to sync again. Also there seems to be a correlation between enabling i2s0 and i2s2 and failure of the eDP link. Thus I disabled i2s0 and i2s2 in my devicetree. It is happening too seldomly to me to pin it to anything yet though.
EDIT: I've pushed an additional commit to my gitlab branch that slows the display clock a tat. While I don't think it will make any difference you are welcome to test it out.
RE: A true mainline Linux Kernel for the Pinebook Pro - danielt - 11-08-2019
(11-08-2019, 06:00 AM)tsys Wrote:(11-08-2019, 04:39 AM)danielt Wrote: I'll go looking for more clues later today but I just wondered if you have seen anything like this yourself as you've been working.
Yeah, interestingly I had one similar event last evening. Upon starting the X user session from lightdm the eDP link failed and was unable to sync again. Also there seems to be a correlation between enabling i2s0 and i2s2 and failure of the eDP link. Thus I disabled i2s0 and i2s2 in my devicetree. It is happening too seldomly to me to pin it to anything yet though.
EDIT: I've pushed an additional commit to my gitlab branch that slows the display clock a tat. While I don't think it will make any difference you are welcome to test it out.
Thanks. I'll do some testing and let you know.
RE: A true mainline Linux Kernel for the Pinebook Pro - tsys - 11-08-2019
Two new feature in the mainline kernel as of now:
- USB-C charging is working (wow, the dt-binding docs of the fusb302 are really shitty)
- Bootup time reduced from more than 3 minutes to less than 30 seconds through new defconfig
RE: A true mainline Linux Kernel for the Pinebook Pro - danielt - 11-08-2019
(11-08-2019, 10:05 AM)tsys Wrote: Two new feature in the mainline kernel as of now:
- USB-C charging is working (wow, the dt-binding docs of the fusb302 are really shitty)
- Bootup time reduced from more than 3 minutes to less than 30 seconds through new defconfig
Reduced boot time will be very welcome. I'll grab that in a second..
I didn't see any benefit from the changed display timings, nor from disabling i2s0 and i2s2. However I chased the error message into the code and came up with a simple DT change to get the driver to tolerate a missing HPD signal. I will need to turn on the debug messages to be absolutely sure I'm still reproducing the original problem but in the mean time if anyone else is having this kind of trouble then the following may help:
Code:
diff --git a/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts b/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts
index 569c1490b024..4df56b48bac4 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts
+++ b/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts
@@ -380,6 +380,7 @@
status = "okay";
pinctrl-names = "default";
pinctrl-0 = <&edp_hpd>;
+ force-hpd;
ports {
edp_out: port@1 {RE: A true mainline Linux Kernel for the Pinebook Pro - tsys - 11-08-2019
(11-08-2019, 01:32 PM)danieltHowever I chased the error message into the code and came up with a simple DT change to get the driver to tolerate a missing HPD signal. I will need to turn on the debug messages to be absolutely sure I\m still reproducing the original problem but in the mean time if anyone else is having this kind of trouble then the following may help: Wrote:Code:diff --git a/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts b/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts
index 569c1490b024..4df56b48bac4 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts
+++ b/arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts
@@ -380,6 +380,7 @@
status = "okay";
pinctrl-names = "default";
pinctrl-0 = <&edp_hpd>;
+ force-hpd;
ports {
edp_out: port@1 {
Nice find! This might indeed be a problem. HPD of the panel is not connected, so this should probably have been in the devicetree all along. Thanks
RE: A true mainline Linux Kernel for the Pinebook Pro - brent.thierens - 11-09-2019
I noticed that the maximum clock is 1.8GHz in the Manjaro build, but the Ubuntu and Debian ones are able to go to 2.02GHz. Can this build be clocked to 2.02GHz as well?
RE: A true mainline Linux Kernel for the Pinebook Pro - tsys - 11-09-2019
(11-09-2019, 04:59 AM)brent.thierens Wrote: I noticed that the maximum clock is 1.8GHz in the Manjaro build, but the Ubuntu and Debian ones are able to go to 2.02GHz. Can this build be clocked to 2.02GHz as well?
Sure. I've added a 2 GHz operating power point @1.3V. One of the next manjaro builds will probably include this change.
RE: A true mainline Linux Kernel for the Pinebook Pro - tsys - 11-09-2019
Major News:
Audio is now working \o/
Was just a simple matter of setting up the iomux correctly
RE: A true mainline Linux Kernel for the Pinebook Pro - sundog - 11-09-2019
(11-09-2019, 09:23 AM)tsys Wrote: Major News:
Audio is now working \o/
Was just a simple matter of setting up the iomux correctly
Awesome!
Preview 2 later today?
RE: A true mainline Linux Kernel for the Pinebook Pro - tsys - 11-09-2019
Just installed the M.2 adapter (after some mechanical optimization) and PCIe works with PCIe 2.0 speed and full 4 lanes.