On 03/31/2016 05:42 AM, Sergey Fedorov wrote: > On 31/03/16 13:48, Alex Bennée wrote: >> I know we are system focused at the moment but does linux-user ever >> flush groups of TBs, say when mappings change? Or does this trigger a >> whole tb_flush? > > Yes, e.g. target_mmap() calls tb_invalidate_phys_range().
Mapping over executable code is exceedingly rare, however. We could do just as well with a full flush, if that's simpler. r~
