Needless to say, it would help if *everyone* would read the I/O
chapter carefully *before* the next meeting, and think about
the outstanding issues and how to resolve them.
Current list of outstanding issues (please send me additional items):
"Discussion" items from the Sept draft:
1. Clarify scope of CAUTIOUS (w.r.t. dynamic processes) [11.1.6]
2. Define end-of-file [11.1.7]
3. Clarify disp & offsets w.r.t. file interoperability [11.1.9]
4. Clarify or delete APPEND mode in OPEN's amode [11.2.1]
5. Clarify "Delete on close" semantics w.r.t. other processes [11.2.3]
6. Clarify scope of "consistency" on CLOSE. [11.2.3]
7. PROPOSAL: allow any size holes in filetype/buftype [11.2.5]
8. PROPOSAL: alternatives to FILE_CONTROL [11.2.8]
9. Clarify RESIZE preallocation error conditions [11.2.11]
10. Clarify RESIZE scope of preallocation [11.2.11]
11. Clarify consistency model at FILE_SYNC [11.2.13]
12. PROPOSAL: Remove *_SHARED_ALL or rename to *_SHARED_ORDERED [11.5.2]
13. PROPOSAL: Move filetype constructors to "Datatype Improvements" [11.6]
14. PROPOSAL: TYPE_SUBARRAY semantics depend on language binding [11.6.3]
Other items:
15. Choose portable offset mechanism: "integer*8" vs. access routines
16. PROPOSAL: Add "count" to VIEW
17. PROPOSAL: Support for non-local I/O services
18. PROPOSAL: Support file interoperability [11.1.9]
19. Define semantics of outstanding requests at CLOSE, ICLOSE, DELETE time
20. PROPOSAL: Add GET_ATOMICITY
21. PROPOSAL: Add GET_FILE_NAME
22. PROPOSAL: Add GET_FILE_MODE
23. Clarify "user must allocate space" for *_GET_* routines
Happy reading,
- bill