On Tue, 29 Jan 2019 at 19:18, Catalin Marinas <catalin.marinas@xxxxxxx> wrote:
>
> On Sun, Jan 27, 2019 at 09:29:42AM +0100, Ard Biesheuvel wrote:
> > Commit 1598ecda7b23 ("arm64: kaslr: ensure randomized quantities are
> > clean to the PoC") added cache maintenance to ensure that global
> > variables set by the kaslr init routine are not wiped clean due to
> > cache invalidation occurring during the second round of page table
> > creation.
> >
> > However, if kaslr_early_init() exits early with no randomization
> > being applied (either due to the lack of a seed, or because the user
> > has disabled kaslr explicitly), no cache maintenance is performed,
> > leading to the same issue we attempted to fix earlier, as far as the
> > module_alloc_base variable is concerned.
> >
> > Note that module_alloc_base cannot be initialized statically, because
> > that would cause it to be subject to a R_AARCH64_RELATIVE relocation,
> > causing it to be overwritten by the second round of KASLR relocation
> > processing.
> >
> > Fixes: f80fb3a3d508 ("arm64: add support for kernel ASLR")
> > Cc: <stable@xxxxxxxxxxxxxxx> # v4.6+
> > Signed-off-by: Ard Biesheuvel <ard.biesheuvel@xxxxxxxxxx>
> > ---
> > arch/arm64/kernel/kaslr.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/arch/arm64/kernel/kaslr.c b/arch/arm64/kernel/kaslr.c
> > index ba6b41790fcd..b09b6f75f759 100644
> > --- a/arch/arm64/kernel/kaslr.c
> > +++ b/arch/arm64/kernel/kaslr.c
> > @@ -88,6 +88,7 @@ u64 __init kaslr_early_init(u64 dt_phys)
> > * we end up running with module randomization disabled.
> > */
> > module_alloc_base = (u64)_etext - MODULES_VSIZE;
> > + __flush_dcache_area(&module_alloc_base, sizeof(module_alloc_base));
>
> Do we need something similar of memstart_offset_seed? If yes, you could
> as well change the returns to a goto out.
>
No, that gets initialized to zero statically, so it isn't affected by this.
