summaryrefslogtreecommitdiff
path: root/open_issues/libpthread_assertion_thread_prevp.mdwn
blob: f93f07d65251a61f441fdad14937ed1c6e6723dc (plain)
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 :)