On 05/04/2016 03:41 PM, Dave Hansen wrote: > But, since we *always* call XSAVES with an instruction mask of -1 and > end up with a requested feature bitmap (RFBM) equal to XCR0, I think we > can do a shortcut because we'll practically *always* have an > xcomp_bv==RFBM==XCR0, which means that all (present) components will > always have an address.
Actually, we depend on that anyway. xstate_comp_offsets[] is completely bogus if we don't have xcomp_bv==RFBM==XCR0. Perhaps we need to add those checks anyway.
