-
Notifications
You must be signed in to change notification settings - Fork 226
/
Bugs
85 lines (58 loc) · 3.02 KB
/
Bugs
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
This file will be a bug track list of issues as I detect them or
people report them, and a sign off that the issue is resolved.
Items marked with an 'x' are 'DONE' and signed off.
x 20080708 1. dtrace -f open
Prove we can see cpu2
Had forced 'cpu = 0' in dtrace_state_buffer() because we
we hitting an assertion.
x 20080708 2. dtrace -f open:entry -- failing probe lookup
Same as apple dtrace. Seems to be a bug in dtrace itself.
x 20080708 3. dtrace -f syscall::open:entry -- failing probe lookup
dtrace itself is broken in the way it parses this. This works
'dtrace -f syscall::open'
20080708 4. wade thru variables lookup
20080708 5. validate ustack/stack
x 20080708 6. fastrap
Merged and compiled, but other bug issues to resolve.
x 20080708 7. fix dtrace -l being erratic on driver reloads
20080708 8. implement hashing on par_alloc.
20080708 9. dtrace -s tests/dt010
hangs system
20080708 10. BUG: soft lockup - CPU#1 stuck for 11s! [dtrace:5862]
Our timer callbacks arent correct so they can hang/kill the
system. (Broken cyclic.c code?)
x 20080708 11. dtrace -s tests/dt011
Invokes CPU_DTRACE_BADADDR from dtrace_copycheck because
we have no base range for the kernel. Use validate_ptr.
Fixed dtrace_copycheck() which assumed Solaris kernelbase
which we dont have. Still get occasional bad pointers
in open() syscalls, but that might be real bugs somewhere!
x 20080712 12. dtrace -s tests/dt013
can cause stack tracing on kernel.
x 20080723 13. kmem_cache_create in dtrace_attach being called twice
or not being freed on driver unload?
x 20080723 14. sys_call_table not visible in 2.6.23.1.
20080729 15. proc2.c: Set pr_dmodel for 64b machine.
x 20081009 16. Build drti.o so that dtrace -G can work.
Done
x 20081010 17. Build /dev/dtrace-helper function needed by dtrace -G.
Done
x 20081019 18. fasttrap_attach() call in fasttrap.c takes us into new
territory which breaks us. Need to init the mutexes in fasttrap_attach.
Done.
20081026 19. dtrace -P fbt
Panics the kernel, but likely due to intercepting stuff we
shouldnt be allowed to.
20090131 20. Need to GC when real procs exit.
20090131 21. Handle fork() syscall to remove or active USDT probes.
20090131 22. 32bit dtrace crashes kernel (or GPFs, doesnt crash)
x 20090204 23. Linux 2.6.9 (AS4) issues
20090215 24. If we unload the driver whilst dtrace is running, panic the kernel.
20090220 25. Intercept process forking -- if fbt tracepoints active in
parent, we need to mimic in the code, else we may have
problems when/if child execs or exits. Code in dtrace
for this, but we havent put the hook for fork.
20090222 26. syscall:sigreturn wants to do things to the stack and
wants a real stack frame. Need to handle that.
20100711 27. syscall::c*: seg fault in kernel
20100711 28. dtrace_xcall: not irq_disable correct.