3 * The tests in tests/ don't apply the grading patches.
5 * We need better and more example programs.
7 - Need an mmap example program as a replacement for the crappy mmap FAQ
10 - How about `diff' and `cmp' programs?
14 - mlfqs tests suck. They aren't even correct, e.g. the amarv
15 submission from win0405 is graded incorrectly.
19 - Don't emphasize that stuff needs to be copied from user space to
20 kernel space. Instead, emphasize validation and suggest that
21 copying is a common solution and that it will be necessary in
22 project 3 and in real OSes. Also revise the grading criteria to
25 - Mark read-only pages as actually read-only in the page table. Or,
26 since this was consistently rated as the easiest project by the
27 students, require them to do it.
29 - Don't provide per-process pagedir implementation but only
30 single-process implementation and require students to implement
31 the separation? This project was rated as the easiest after all.
32 Alternately we could just remove the synchronization on pid
33 selection and check that students fix it.
37 - Discuss the perils of mixing dirty bits between kernel and user virtual
42 - Update grading/vm to reflect new mmap, munmap forms.
46 - Increase maximum disk size from 8 MB to something that actually
47 requires doubly indirect nodes. There is a negative pressure here
48 from the bitmap object--perhaps we need a specialized bitmap that
49 doesn't have to be all in-memory at once.
51 Alternatively, shrink the inode size.
53 - Add mkdir and ls example user programs.
55 - Add option to disable buffer cache.
57 - Get rid of "dump" commands--they're not really useful.
61 - Need a better way to measure performance improvement of buffer
62 cache. Some students reported that their system was slower with
63 cache--likely, Bochs doesn't simulate a disk with a realistic
66 - Clarify effect of remove(cwd).
70 - Finish writing tour.
72 - Add "Digging Deeper" sections that describe the nitty-gritty x86
73 details for the benefit of those interested.
75 - Add explanations of what "real" OSes do to give students some
82 . Low-level x86 stuff, like paged page tables.
84 . Specifics on how to implement sbrk, malloc.
88 . opendir/readdir/closedir
90 . everything needed for getcwd()
92 - Add src/testcases/vm, src/testcases/filesys and make it clear to use
97 - Release some of them.
99 - The threads, userprog, vm test source files could use
100 factorization and cleanup along the lines of fslib in the filesys
103 - The p1-4.c testcase needs significant tuning. Currently it takes
104 too long (especially when SHOW_PROGRESS is turned on) and doesn't
105 show significant improvement.
109 - Make threads test use a program and arguments like the other
112 - Revise command-line arguments for sensibility instead of for
113 Nachos compatibility.
115 - Get rid of all compile-time flags, e.g. MLFQS should be a runtime
118 - Need an optimization barrier and an explanation of it in the
121 - Need to check the wait system call more thoroughly.
123 - Library is not precisely standard, e.g. ptrdiff_t should not be