Rusty Russell wrote:
> Actually, your paravirt_patch_insns has similar logic anyway, so this
> code could collapse (it should fall back to paravirt_patch_default tho
> IMHO).
>   

I wanted to get rid of the table because its now sparse, and possibly
fairly large (since the special save flags & disable is at offset
0x80).  And a switch would be pretty clean anyway.

How does this look?  Compiles, but untested.

    J