Hi Ard, On Tue, Oct 2, 2018 at 5:54 AM Ard Biesheuvel <ard.biesheu...@linaro.org> wrote: > So Arnd's suggestion to switch to armv3-m would actually be feasible > then? The code in question does not use ldrh only umull, and so it > should build with armv3m as well. And we will even generate some > better code for RiscPC if we apply it to both the cflags and the > asflags.
It turns out armv3m isn't a solution for the crypto code, since it does other armv4-isms. So we'll need to find another solution for that; I'm partial to my initial Kconfig one just doing "depends !CPU_32V3". With regards to RiscPC though, given that armv3 codegen might be better, it sounds like it's a good idea to set -march=armv3m anyway for the platform, irrespective of fixing the problem at hand here. Regards, Jason
