<hexdump0815>
jenneron[m]: i was not able to figure out where you set the kernel cmdline in pmos on veyron, but maybe make sure you have the "gpt" kernel cmdline option set in order to make a kpart kernel work well with emmc
<hexdump0815>
i think there are also patches to work around the strange emmc partition setup on veyron and maybe pmos is using those - then this might not be relevant for pmos
<hexdump0815>
background is: some veyrons seem to not work well with my images due to u-boot problems (booting often ends in a blank screen after u-boot and only from time to time towrks well)
<hexdump0815>
those work well when used with writing the kernel directly to the kernel partition, but at least my kernels missed the "gpt" kernel option to take into accound the secondary gpt partition as the primary seems to be corrupted by design by default on veyron
<hexdump0815>
so my kernels this way onlyworked with sd, but not with emmc
<hexdump0815>
as said, maybe all this is not relevant for pmos, but just wanted to let you know just in case ...
<jenneron[m]>
3) we don't use u-boot here, we use depthcharge
<hexdump0815>
jenneron[m]: thanks for the link - yes that patch i ment ... i actually looked at that deviceinfo file, but must have overlooked the cmdline as it is so short - i looked around for something longer :)
<hexdump0815>
i know that pmos is using depthcharge - at some point i'll look into switching to it as well on veyron as the u-boot seems to be quite broken for some devices
<jenneron[m]>
yeah
<jenneron[m]>
i found the easiest to just use depthcharge on all of them
<jenneron[m]>
including exynos
<hexdump0815>
i'm not aware of that problem from the issue - when i was daily drive testing my veyron i had usb kdb and mouse connected and used suspend/resume without any problem - that was with an early v6.1