On Fri, 2024-03-22 at 17:36 -0700, Linus Torvalds wrote: > On Fri, 22 Mar 2024 at 17:21, Kent Overstreet > <kent.overstreet@xxxxxxxxx> wrote: > > > > Besides that there's cross arch support to think about - it's hard > > to > > imagine us ever ditching our own atomics. > > > [... SNIP ...] > > > > I was thinking about something more incremental - just an optional > > mode > > where our atomics were C atomics underneath. It'd probably give the > > compiler people a much more effective way to test their stuff than > > anything they have now. > > I suspect it might be painful, and some compiler people would throw > their hands up in horror, because the C++ atomics model is based > fairly solidly on atomic types, and the kernel memory model is much > more fluid. > > Boqun already mentioned the "mixing access sizes", which is actually > quite fundamental in the kernel, where we play lots of games with > that > (typically around locking, where you find patterns line unlock > writing > a zero to a single byte, even though the whole lock data structure is > a word). And sometimes the access size games are very explicit (eg > lib/lockref.c). > > But it actually goes deeper than that. While we do have "atomic_t" > etc > for arithmetic atomics, and that probably would map fairly well to > C++ > atomics, in other cases we simply base our atomics not on _types_, > but > on code. > > IOW, we do things like "cmpxchg()", and the target of that atomic > access is just a regular data structure field. > > It's kind of like our "volatile" usage. If you read the C (and C++) > standards, you'll find that you should use "volatile" on data types. > That's almost *never* what the kernel does. The kernel uses > "volatile" > in _code_ (ie READ_ONCE() etc), and uses it by casting etc. > > Compiler people don't tend to really like those kinds of things. Just for my understanding: Why don't they like it? I guess since compiler people have to support volatile pointers anyways, temporarily casting something to such a volatile pointer shouldn't be a problem either – so they don't dislike it because it's more difficult to implement, but because it's more difficult to verify for correctness? P. > > Linus >
