On Tue, Oct 10, 2023 at 03:10:54PM +0200, Borislav Petkov wrote: > On Tue, Oct 10, 2023 at 12:10:56PM +0200, Peter Zijlstra wrote: > > Now, obviously you really don't want boot_cpu_has() in > > __VIRTUAL_MASK_SHIFT, that would be really bad (Linus recently > > complained about how horrible the code-gen is around this already, must > > not make it far worse). > > You mean a MOV (%rip) and a TEST are so horrible there because it is > a mask? > > I'd experiment with it when I get a chance... That gets you a memory-reference and potential cachemiss what should have been an immediate :/
