[Mono-bugs] [Bug 75034][Nor] Changed - Gecko# segfault not converted to NullRef

bugzilla-daemon at bugzilla.ximian.com bugzilla-daemon at bugzilla.ximian.com
Tue May 24 20:21:26 EDT 2005


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 jonpryor at vt.edu.

http://bugzilla.ximian.com/show_bug.cgi?id=75034

--- shadow/75034	2005-05-24 18:35:01.000000000 -0400
+++ shadow/75034.tmp.19557	2005-05-24 20:21:26.000000000 -0400
@@ -3,21 +3,21 @@
 Version: unspecified
 OS: unknown
 OS Details: 
 Status: REOPENED   
 Resolution: 
 Severity: Unknown
-Priority: Wishlist
+Priority: Normal
 Component: misc
 AssignedTo: mono-bugs at ximian.com                            
 ReportedBy: alexmipego at hotmail.com               
 QAContact: mono-bugs at ximian.com
 TargetMilestone: ---
 URL: 
 Cc: 
-Summary: gtk# segfault not converted to NullRef
+Summary: Gecko# segfault not converted to NullRef
 
 Using the testcase I attach Mono (after 5 seconds) dies with segmentation
 fault. The gui isn't thread-safe, so thats a buggy code, yet Mono should
 handle this and throw an Exception, not a segmentation fault.
 
 ------- Additional Comments From alexmipego at hotmail.com  2005-05-24 08:57 -------
@@ -45,6 +45,22 @@
 
 ------- Additional Comments From bmaurer at users.sf.net  2005-05-24 18:35 -------
 Can we prove that the segfault was really a nullref?
 
 Btw, trying to catch this exception would be horrible. God only knows
 hot the gui libs have been corrupted.
+
+------- Additional Comments From jonpryor at vt.edu  2005-05-24 20:21 -------
+First, it's Gecko# that shows up in the backtrace as causing the segfault.
+
+Second, yes, only God knows what memory Gecko# has corrupted.
+
+However, for *any* external library invoked through P/Invoke, God
+knows what memory may have been corrupted, but mono *still* converts
+the SIGSEGV into a NullReferenceException.  Consider:
+
+    Mono.Unix.Stdlib.fclose (System.IntPtr.Zero);
+
+This causes a SIGSEGV (fclose(3) doesn't like a NULL pointer), but
+mono generates a NullReferenceException + stack trace for it.
+
+Why shouldn't Gecko# be any different?


More information about the mono-bugs mailing list