> I will commit the original one and review this one in a sec. I committed the patch as r192823. On the new one, I think my only comment is the same as Jim Grosbach's in the other patch. If we know the size of the instruction, we should skip that, not just one byte. Cheers, Rafael