On Tue, Feb 11, 2020 at 02:28:06PM -0800, Andy Lutomirski wrote: > On Tue, Feb 11, 2020 at 5:53 AM Joerg Roedel <joro@xxxxxxxxxx> wrote: > It would be nice if this could reuse the existing in-kernel > instruction decoder. Is there some reason it can't? It does, see patch 5, which makes the inat-tables generator script suitable for pre-decompression boot code. Actually every instruction-caused #VC exception will decode the instruction to get its length. Regards, Joerg
