On Tue, 14 Jul 2020 10:31:25 +0300 Boris Pismenny wrote: > Now we have an ASIC that uses this API, and I'd like to show the best > possible outcome, and not the best possible given an arbitrary > limitation that avoids an error where the user does something > erroneous.
I would not call correctness an arbitrary limitation. AFAIU page cache is shared, one application thinking that files it opens can't be modified is no guarantee that the pages themselves will remain unchanged. Isn't support for read-only page cache entries also a problem for RDMA when it comes to fs checksums? Sounds like a problem area that needs real solutions.
