On Tue, 24 Feb 2015, Leonid Yegoshin wrote: > > For simplicity perhaps on SMP we should just always use hit operations > > regardless of the size requested. > > High performance folks may not like doing a lot of stuff for 8MB VMA release > instead of flushing 64KB. What kind of a use case is that, what does it do? > Especially taking into account TLB exceptions and postprocessing in > fixup_exception() for swapped-out/not-yet-loaded-ELF blocks. The normal use for cacheflush(2) I know of is for self-modifying or other run-time-generated code, to synchronise caches after a block of machine code has been patched in -- SYNCI can also be used for that purpose these days, but the syscall long predates the presence of this instruction in the architecture . I fail to see any other need to force data out of cache in a user program (apart from also the corner case of benchmarking cold caches, such as in Zenon's case, but there the performance of the preparatory cache invalidation does not really matter). Why would you expect VM to have been swapped out or not loaded where data has just been written? Why would you want to call cacheflush(2) for blocks of data that haven't been written? Maciej
