On Mon, May 14, 2018 at 10:46:33AM +0100, Mark Rutland wrote: > We can zero GPRs x0 - x29 upon entry from EL0 to make it harder for > userspace to control values consumed by speculative gadgets. > > We don't blat x30, since this is stashed much later, and we'll blat it > before invoking C code. > > Signed-off-by: Mark Rutland <mark.rutland@xxxxxxx> > Cc: Catalin Marinas <catalin.marinas@xxxxxxx> > Cc: Will Deacon <will.deacon@xxxxxxx> > --- > arch/arm64/kernel/entry.S | 9 +++++++-- > 1 file changed, 7 insertions(+), 2 deletions(-) > > diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S > index 13afefbf608f..4dd529fd03fd 100644 > --- a/arch/arm64/kernel/entry.S > +++ b/arch/arm64/kernel/entry.S > @@ -62,6 +62,12 @@ > #endif > .endm > > + .macro clear_gp_regs > + .irp n,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29 > + mov x\n, xzr > + .endr > + .endm > + Looks OK, but consider moving _for from fpsimdmacros.h to assembler.h and just writing _for n, 0, 29, mov x\n, xzr (could even omit the wrapper macro, since this is a one-liner). The implementation of _for is a bit gross, but since we already have it, we might as well use it. [...] Cheers ---Dave
