[Mono-bugs] [Bug 58065][Maj] Changed - Exception handling needs to be moved out of signal handlers

[bugzilla-daemon@bugzilla.ximian.com]

Please do not reply to this email- if you want to comment on the bug, go to the
URL shown below and enter your comments there.

Changed by dick@ximian.com.

http://bugzilla.ximian.com/show_bug.cgi?id=58065

--- shadow/58065	2004-05-27 14:03:22.000000000 -0400
+++ shadow/58065.tmp.9309	2004-06-18 05:17:18.000000000 -0400
@@ -10,13 +10,13 @@
 Component: misc
 AssignedTo: mono-bugs@ximian.com                            
 ReportedBy: umuff@quark.com               
 QAContact: mono-bugs@ximian.com
 TargetMilestone: V1
 URL: 
-Summary: Random Crash or Freeze running threadpool.exe
+Summary: Exception handling needs to be moved out of signal handlers
 
 Running make install in mono/runtime will result in either a failure with thread_info failed error, 
 or a freeze, here is the info for a freeze:
 
 (gdb) i threads
   2 process 5184 thread 0x1103  0x010b6900 in _wapi_handle_get_shared_segment 
@@ -518,6 +518,20 @@
 at object.c:1419
 #7  0x08066535 in mono_main (argc=1, argv=0xbfffefd4) at driver.c:801
 #8  0x08065473 in main (argc=2, argv=0xbfffefd4) at main.c:6
 #9  0x42015704 in __libc_start_main () from /lib/tls/libc.so.6
 #0  0xffffe002 in ?? ()
 
+
+------- Additional Comments From dick@ximian.com  2004-06-18 05:17 -------
+I think the original deadlock reported by Urs is now fixed in CVS.
+
+The traces from Miguel and Paolo I believe are to do with running the
+Thread.Abort() (and other) exception in a signal handler.  Note the
+sigusr1_signal_handler frame in almost every thread of Paolo's.  Ditto
+sigfpe_signal_handler in Miguel's.
+
+I can run threadpool.exe continuously for hours on end on an SMP
+machine if I disable the Thread.Abort() pthread_kill().
+
+I'm renaming this bug to reflect this info.
+