172464 - The switch test hangs on HP-UX 11.00, 32-bit optimized build

Status: RESOLVED WONTFIX

(NSPR :: NSPR, defect)

Description

[Wan-Teh Chang]

Version: NSPR 4.2.2

The switch test hangs on HP-UX 11.00, 32-bit optimized
build.  The test machine is sbshp1.

The thread stacks are:

(gdb) info threads
  2 system thread 23357  warning: reading register r3: No data
0xc01eb408 in __ksleep () from /usr/lib/libc.2
* 1 system thread 7054  warning: reading register r3: No data
0xc0fe2380 in ___lwp_wait_sys ()
   from /usr/lib/libpthread.1
(gdb) thread 1
[Current thread is already 1 (system thread 7054)]
(gdb) where
#0  0xc0fe2380 in ___lwp_wait_sys () from /usr/lib/libpthread.1
#1  0xc0fe1fd0 in _lwp_wait () from /usr/lib/libpthread.1
#2  0xc0fdb0bc in __vp_join () from /usr/lib/libpthread.1
#3  0xc0fd9d70 in pthread_join () from /usr/lib/libpthread.1
#4  0xc1378ab8 in PR_JoinThread ()
   from /tmp_mnt/share/builds/sbsrel2/nspr/nspr422/builds/20021003.1/build_tree_
ux/hpux32.opt/pr/tests/../../dist/lib/libnspr4.sl
#5  0xce38 in Switch ()
#6  0xc136737c in PR_Initialize ()
   from /tmp_mnt/share/builds/sbsrel2/nspr/nspr422/builds/20021003.1/build_tree_
ux/hpux32.opt/pr/tests/../../dist/lib/libnspr4.sl
#7  0xd2e0 in main ()
(gdb) thread 2
warning: reading register r3: No data
[Switching to thread 2 (system thread 23357)]
#0  0xc01eb408 in __ksleep () from /usr/lib/libc.2
(gdb) where
#0  0xc01eb408 in __ksleep () from /usr/lib/libc.2
#1  0xc0fd81bc in _lwp_cond_timedwait () from /usr/lib/libpthread.1
#2  0xc0fd7f0c in pthread_cond_wait () from /usr/lib/libpthread.1
#3  0xc137033c in PR_WaitCondVar ()
   from /tmp_mnt/share/builds/sbsrel2/nspr/nspr422/builds/20021003.1/build_tree_
ux/hpux32.opt/pr/tests/../../dist/lib/libnspr4.sl
#4  0xcad8 in Notified ()
#5  0xc13782a8 in _pt_root ()
   from /tmp_mnt/share/builds/sbsrel2/nspr/nspr422/builds/20021003.1/build_tree_
ux/hpux32.opt/pr/tests/../../dist/lib/libnspr4.sl
#6  0xc0fd9c5c in __pthread_create_system () from /usr/lib/libpthread.1
(gdb)

Note that the switch test calls PR_Interrupt, so this
hang could be caused by bug 165586.

Comment 1

[hajma]

I'm also observing a the switch test to hang indefinitely on Solaris 11.4, both x86 and SPARC. NSPR 4.29

pstack 7217

7217: ./switch
fec5d544 lwp_park (0, 0, 0)
fec56918 cond_wait_queue (24908, 248a0, 0, 0, 0, 0) + 4c
fec56f54 cond_wait (24908, 248a0, fecb8940, 2946, 2800, ff0e2a40) + 10
fec56f94 pthread_cond_wait (24908, 248a0, 24bd8, ffbff3a8, ffffffff, 0) + 8
ff11ae08 PR_WaitCondVar (24900, ffffffff, 24bb0, 2, 24900, 23e90) + c8
000119fc Switch (b, 23000, 1, 24bb0, 64, 230a0) + 1fc
ff1111b0 PR_Initialize (11800, 1, ffbff55c, 0, 0, 12c14) + 10
00011454 _start (0, 0, 0, 0, 0, 0) + 5c

Comment 2

[BugBot [:suhaib / :marco/ :calixte]]

The bug assignee didn't login in Bugzilla in the last 7 months.
:KaiE, could you have a look please?
For more information, please visit auto_nag documentation.