On 4/25/19 9:04 AM, Palmer Dabbelt wrote: > c.lwsp with rd=0 should be an illegal instruction, but it's not. We'd need to > re-introduce the shim to handle that quirk, but since it's not actually a new > bug I'm OK taking the patch set and fixing that later.
That should be fixable in the decode file, without a shim. See patch 5, where I introduce trans_illegal, for doing exactly this for c.addi4spn. r~
