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
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
|
[[!meta copyright="Copyright © 2011, 2013 Free Software Foundation, Inc."]]
[[!meta license="""[[!toggle id="license" text="GFDL 1.2+"]][[!toggleable
id="license" text="Permission is granted to copy, distribute and/or modify this
document under the terms of the GNU Free Documentation License, Version 1.2 or
any later version published by the Free Software Foundation; with no Invariant
Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license
is included in the section entitled [[GNU Free Documentation
License|/fdl]]."]]"""]]
[[!meta title="libpthread: __pthread_enqueue: Assertion `thread->prevp == 0'
failed"]]
[[!tag open_issue_libpthread]]
# IRC, OFTC, #debian-hurd, 2011-10-21
[Python testsuite]
<pinotree> [169/340/1] test_logging
<pinotree> python:
/home/pino/sources/hurd/hurd-20110519/./libpthread/pthread/pt-internal.h:109:
__pthread_enqueue: Assertion `thread->prevp == 0' failed.
<pinotree> sigh
## IRC, freenode, #hurd, 2011-10-21
<pinotree> am i missing anything, or in libpthread the __pthread_threads
list does not ever has elements removed from it?
<pinotree> ... thus potentially causing
"./libpthread/pthread/pt-internal.h:109: __pthread_enqueue: Assertion
`thread->prevp == 0' failed." because threads can be appended on
__pthread_dealloc() to the __pthread_free_threads list as well?
<pinotree> maybe reusing the same next+prevp pointers in the __pthread
struct for more than one list at the same time isn't a good idea...
## IRC, freenode, #hurd, 2011-10-23
<youpi> pinotree: I don't understand the relation between thread->prevp !=
0 and the __pthread_threads list
<youpi> the list never has elements removed indeed, since libpthread never
frees __pthread structures apparently
<pinotree> youpi: ye sorry, that relation is indeed nonsense
<youpi> in which condition did you get prevp != 0
<pinotree> i wa trying to find some explaination for the "thread->prevp ==
0" assertion in the _queue function
<youpi> ?
<pinotree> *was
<youpi> it's not obvious to me how libpthread makes sure the various
cond/mutex/rwlock make sure that it's not queued several times
<pinotree> yeah
<pinotree> apparently prevp/next are used for lists of held
waitcond/mutex/rwlock and free threads
# IRC, freenode, #hurd, 2013-03-20
<braunr> aw
<braunr> i hit the ext2fs.static: ./pthread/pt-internal.h:122:
__pthread_enqueue: Assertion `thread->prevp == 0' failed.
<braunr> assertion
<braunr> looks like there is a deadlock on assert
<braunr> which might explain why i never saw progress when i tested that in
the past
## IRC, freenode, #hurd, 2013-04-21
<braunr> damn, there still bugs in libpthread
<braunr> (about prevp not being null when it should i mean)
<pinotree> braunr: found another trigger for that?
<braunr> no
<braunr> it's so random i wonder if it's not a completely unrelated
corruption
<braunr> pinotree: also, i'm having more of these issues with my custom
hurd packages that let threads exit after some time from managing ports
<braunr> (i removed the libports_stability patch)
<braunr> i once had this : http://www.sceen.net/~rbraun/darnassus_crash.png
[The assertion failure.]
## IRC, freenode, #hurd, 2013-04-23
<braunr> removing the libports_stability patch exposed bugs in libpthread,
triggering assertions when queueing/dequeue threads from a queue (but i
don't know which one / in which function)
## IRC, freenode, #hurd, 2013-06-25
<pinotree> braunr:
https://buildd.debian.org/status/fetch.php?pkg=libmemcached&ver=1.0.17-2&arch=hurd-i386&stamp=1372165732
<pinotree> make: ./pthread/pt-internal.h:122: __pthread_enqueue: Assertion
`thread->prevp == 0' failed. \o/
<pinotree> (it should rather be /o\, but better pretend not)
<braunr> pinotree: yes, we regularly see it
<braunr> pinotree: how long has the machine been running at this point ?
<pinotree> dunno, you should ask samuel about that
<pinotree> does it happen after N hours/days?
<braunr> a few days of moderate to high activity yes
<pinotree> ah ok
<braunr> and i actually see this error much more often when i disable the
libports stability patch in the hurd debian package
<braunr> so i guess something is wrong with thread recycling
<braunr> but i wanted to completely rewrite that part with the new kernel
call i asked bddebian to work on :)
|
