On 13/08/2015 17:32, Peter Maydell wrote: >> > TBs are physically-tagged so invalidates are not too frequent unless the >> > guest does self-modifying code or swaps. They shouldn't be a source of >> > tb_lock contention except if you do SMC (not just dynamic recompilation: >> > really self-modifying code). > Do guests with writeable data and code in the same page also > get treated as SMC for these purposes? (That's more common > than real SMC.)
Not after a short while, when the code_bitmap kicks in. Paolo