1. Lightweight Remote Procedure Call Brian N. Bershad, Thomas E. Anderson, Edward D. Lazowska, Henry M. Levy ACM Transactions Vol. 8, No. 1, February 1990, pp. 37-55 presented by Ian Dees for PSU CS533, Jonathan Walpole February 2, 2009

2. sharing resources inside the kernel

3. “protection domains” like address spaces; may or may not be address spaces

4. traditional approaches capability systems vs. remote procedure call

5. remote procedure call inspired by distributed computing separate address spaces coarse-grained access messages wrapped in stubs and proxies

6. applying RPC ideas locally RPC model “appropriate for managing subsystems, even those not primarily intended for remote operation” but how to make message passing efficient?

7. naïve approach treat local as a special case of remote treat atomic parameters as a special case of structured data

8. instead, make the common case fast

9. most calls are local even in distributed systems

11. most parameters are simple even in complex APIs

13. sources of overhead stubs message copying sender validation message queueing thread scheduling context switching thread dispatch

14. sender validation when should the kernel check permissions?

15. (capability systems) ‏ single address space for parameter passing fine-grained: token ≈ pointer + permissions principle of least privilege

16. analogies in user space FILE* f = fopen("path/to/file", "r"); LPSECURITY_ATTRIBUTES

17. advantages / disadvantages + eliminate many stability / security pitfalls - cumbersome to program - difficult to implement efficiently

18. existing optimizations memory mapping handoff scheduling global message buffers

19. LRPC’s four techniques

20. 1. simple control transfer look up addresses once, at discovery time (clerks and bindings) ‏ server uses caller’s thread return directly to caller cache execution stacks for reuse

21. 2. simple data transfer map argument stacks into both spaces aggressively reduce copies (but call-by-reference involves an extra copy) ‏ fall back on marshaling

22. 3. simple stubs server stub branches straight to procedure generate simple assembly for common cases exceptions, large parameters fall back to RPC

23. 4. design for concurrency distribute domains across processors monitor concurrency misses and wake CPUs use fine-grained locking of argument stacks

24. edge cases remote vs. local activation stack size limits domain termination

25. results