I see the problem with 4.7-rc6. Is there a different kernel I should test with?
On Fri, Jul 22, 2016 at 4:39 PM, Tom Rini <tr...@konsulko.com> wrote: > On Fri, Jul 22, 2016 at 10:40:11AM -0400, Sandy Patterson wrote: > > > Commit c09d29057a and df120142f36 break kernel loading on rock2 board. > > > > console output: > > Starting kernel ... > > > > ### ERROR ### Please RESET the board ### > > > > This reverts commit c09d29057ab0b04db0857d319c6bff74de31b9c3. > > > > Conflicts: > > arch/arm/cpu/armv7/cache_v7.c > > arch/arm/cpu/armv7/cache_v7_asm.S > > > > Signed-off-by: Sandy Patterson <apatter...@sightlogix.com> > > So, we cannot do this. The problem is two-fold. First, newer GCCs will > cause failure to boot with the way the code was, which is how we found > this problem. The second is that after some offline conversations with > ARM Ltd people (and some other smart folks too, these operations cannot > safely be done in C, there's just no way. So we borrowed what the > kernel does. You need to figure out what behavior the (old?) kernel is > relying on and either emulate that with a flag (like sunxi does, or at > least did, at one point for some other issues of correct behavior vs > existing kernels) or just not support those older kernels. > > -- > Tom > _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
