On 2017-08-03, Cyril Brulebois wrote: > d-i now FTBFSes on armhf, due to: > ,---[ hd-media ]--- > | gen-hd-image: Installing /usr/lib/u-boot/firefly-rk3288/u-boot-spl.rksd at > sector 64 ... > | gen-hd-image: Installing /usr/lib/u-boot/firefly-rk3288/u-boot.img at > sector 256 ... > | config/armhf//hd-media.cfg:33: recipe for target > 'hd-media_images_concatenateable' failed > | make[2]: *** [hd-media_images_concatenateable] Error 1 ... > ,---[ u-boot ]--- > | Providing u-boot binaries for Firefly-RK3288 ... > | cp: cannot stat '/usr/lib/u-boot/firefly-rk3288/u-boot.img': No such file > or directory > | config/armhf//u-boot.cfg:8: recipe for target 'u-boot-binaries' failed > `--- > > I suppose this is due to this change in u-boot on 2017-08-01: > | u-boot (2017.07+dfsg1-2) unstable; urgency=medium > | . > | * u-boot-rockchip: > | - Ship u-boot.bin in firefly-rk3288 instead of u-boot.img. > | - Add NEWS file explaining the change for firefly-rk3288. > > See https://tracker.debian.org/news/860117 > > It'd be nice if someone could double check what new path is needed, and > push a commit to master after a test build. Otherwise I'll do so in a > couple of days.
I'm sorry, my mistake, I totally forgot about u-boot's impact on debian-installer again... *sigh* This may actually require changing the d-i code(the new method requires appending two things together before dd'ing them, rather that dd'ing two things at specific locations), or more changes to u-boot (I could pregenerate that part in u-boot, though that means shipping redundant bits). Might be best to temporarily disable the firefly-rk3288 in d-i until I figure out what's best to do... live well, vagrant
signature.asc
Description: PGP signature
