Build regressions/improvements in v4.2-rc8

[Alexander Graf]

On 24.08.15 10:36, Geert Uytterhoeven wrote:
> On Mon, Aug 24, 2015 at 10:34 AM, Geert Uytterhoeven
> <geert at linux-m68k.org> wrote:
>> JFYI, when comparing v4.2-rc8[1] to v4.2-rc7[3], the summaries are:
>>   - build errors: +4/-7
> 
> 4 regressions:
>   + /home/kisskb/slave/src/include/linux/kvm_host.h: error: array
> subscript is above array bounds [-Werror=array-bounds]:  => 430:19
>     (arch/powerpc/kvm/book3s_64_mmu.c: In function 'kvmppc_mmu
> _book3s_64_tlbie':)
> 
> powerpc-randconfig (seen before in a v3.15-rc1 build?)

I'm not quite sure what's going wrong here. The code in question does

  kvm_for_each_vcpu(i, v, vcpu->kvm)
    kvmppc_mmu_pte_vflush(v, va >> 12, mask);

and IIUC the thing we're potentially running over on would be
kvm->vcpus[i]. But that one is bound by the kvm_for_each_vcpu loop, no?


Alex