| Summary: | StatusWidget causes frysk to hang on garbage collection in FC5 - gc deadlock | ||
|---|---|---|---|
| Product: | frysk | Reporter: | Adam Jocksch <ajocksch> |
| Component: | general | Assignee: | Unassigned <frysk-bugzilla> |
| Status: | NEW --- | ||
| Severity: | normal | CC: | dnovillo |
| Priority: | P2 | ||
| Version: | unspecified | ||
| Target Milestone: | --- | ||
| Host: | Target: | ||
| Build: | Last reconfirmed: | ||
| Bug Depends on: | 2304 | ||
| Bug Blocks: | 1839, 1664, 2288 | ||
|
Description
Adam Jocksch
2006-02-10 20:37:59 UTC
The StatusWidget creates an extra thread vis:
Thread 5 (Thread -1210287184 (LWP 8867)):
#0 0x00c6e402 in ?? ()
#1 0x005127a6 in ?? () from /lib/libpthread.so.0
#2 0x02745213 in _Jv_CondWait (cv=0x304bc00, mu=0x304bc20, millis=0x0,
nanos=0x0) at ../../../libjava/posix-threads.cc:166
#3 0x02722d81 in gnu::gcj::runtime::FinalizerThread::run (this=0x4c348)
at ../../../libjava/gnu/gcj/runtime/natFinalizerThread.cc:57
#4 0x02733209 in _Jv_ThreadRun (thread=0x4c348)
at ../../../libjava/java/lang/natThread.cc:296
(More stack frames follow...)
Thread 4 (Thread -1220777040 (LWP 8868)):
#0 0x00c6e402 in ?? ()
#1 0x00407a5c in *__GI___poll (fds=0x46bff4, nfds=0x7, timeout=0xffffffff)
at ../sysdeps/unix/sysv/linux/poll.c:86
#2 0x002e744c in g_main_context_check () from /usr/lib/frysk/libglib-2.0.so.0
#3 0x002e7957 in g_main_loop_run () from /usr/lib/frysk/libglib-2.0.so.0
#4 0x008043de in link_thread_io_context () from /usr/lib/libORBit-2.so.0
(More stack frames follow...)
Thread 3 (Thread -1221313616 (LWP 8869)):
#0 0x00c6e402 in ?? ()
---Type <return> to continue, or q <return> to quit---
#1 0x00407a5c in *__GI___poll (fds=0x46bff4, nfds=0x0, timeout=0x10b8)
at ../sysdeps/unix/sysv/linux/poll.c:86
#2 0x080fbb8d in ?? ()
#3 0x080f6f4f in ?? ()
#4 0x080f7193 in ?? ()
(More stack frames follow...)
Thread 2 (Thread -1232794704 (LWP 8870)):
#0 0x00cfd567 in timer_helper_thread (arg=0x0)
at ../nptl/sysdeps/unix/sysv/linux/timer_routines.c:74
#1 0x00510b80 in start_thread (arg=0xb6850bb0) at pthread_create.c:261
#2 0x004119ce in ?? () from /lib/libc.so.6
Thread 1 (Thread -1208162624 (LWP 8866)):
#0 0x00c6e402 in ?? ()
#1 0x00407a5c in *__GI___poll (fds=0x46bff4, nfds=0x5, timeout=0xffffffff)
at ../sysdeps/unix/sysv/linux/poll.c:86
#2 0x002e744c in g_main_context_check () from /usr/lib/frysk/libglib-2.0.so.0
#3 0x002e7957 in g_main_loop_run () from /usr/lib/frysk/libglib-2.0.so.0
#4 0x00a24fc1 in gtk_main () from /usr/lib/frysk/libgtk-x11-2.0.so.0
(More stack frames follow...)
suspect it is causing a garbage collect hang.
adam and I examing strace shows the sequence: - successfull garbage collect ... - creation of real-time timer manager thread (I'm guessing that is what it is) - timer_create - creation of short-lived timer expired thread; then exit - repeat; in some cases several exist in parallel; but all appear to exit then - garbage collect sigpwr signals things hang; suggests bug is bad interaction with the timer event and the garbage collector. Upstream fedora tracker: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=180637 |