03:21 <CounterPillow> Kwiboo: are you working on PineTab2 bootloader stuff?

05:52 <CounterPillow> ok I see that yes you are :) https://github.com/Kwiboo/u-boot-rockchip/commit/66855bea7fc388b938e958b2039e67e704d10040

07:50 <Kwiboo> CounterPillow: yes, was testing u-boot on my rk356x devices this past weekend and got a delivery notice on the PT2, so I prepared a u-boot defconfig for it :-)

07:51 <Kwiboo> turned out it needed similar workaround as quartz64-a in regard to sdmmc pwren pin

07:53 <Kwiboo> also managed to test pcie/ahci and the sata port on the odroid-m1, so now some defconfigs are updated to support that

07:53 <CounterPillow> nice!

07:54 <Kwiboo> asonix testing of u-boot on PT2 also uncovered a possible bug related to relative paths in bootstd extlinux handling, and think I may have found the source of this issue

07:56 <Kwiboo> also possible that there is an issue with case sensitivity in resolving efi/boot/bootaa64.efi using bootstd efi

07:56 <CounterPillow> EFI partitions are usually vfat, right? Does that not do casefolding?

07:57 <Kwiboo> yes that was my understanding, so very confused

07:57 <CounterPillow> (though technically they can be anything, vfat is just the minimum)

07:57 <Kwiboo> my freebsd 14 image was discovered, but not my fedora server 38, and was sure that it was discovered when I tested few weeks back

07:57 <CounterPillow> Hmmmm

08:00 <Kwiboo> at least my reply on mailing list indicated that it was working: https://patchwork.ozlabs.org/project/uboot/patch/20230506020305.1371008-2-sjg@chromium.org/#3108597

08:01 <Kwiboo> that was on a rk3399 so maybe something differs with rk356x

08:54 <Kwiboo> hum, re-testing on a rock-3a did not show the same issues with regard to extlinux relative path or failing to locate bootaa64.efi for debian/fedora/freebsd, will need to re-test on my pt2 later