On 9/24/2010 4:17 AM, Corinna Vinschen wrote:
Can you revert to the latest from CVS and try again with this patch applied:
Index: autoload.cc
===================================================================
RCS file: /cvs/src/src/winsup/cygwin/autoload.cc,v
retrieving revision 1.174
diff -u -p -r1.174 autoload.cc
--- autoload.cc 23 Sep 2010 20:18:16 -0000 1.174
+++ autoload.cc 24 Sep 2010 09:15:40 -0000
@@ -233,7 +233,7 @@ std_dll_init ()
dll->handle = h;
}
else if (!(func->decoration& 1))
- api_fatal ("could not load %W, %E", dll->name);
+ api_fatal ("could not load %W, %E", dll_path);
else
dll->handle = INVALID_HANDLE_VALUE;
}
If the error occurs again, what is the path printed in the error
message? Is it sane? Does the directory correspond to your local
X:\Windows\System32 directory?
Done. The crash came back and here's the result:
1 [main] bclanc 2544! C:\budcat\tools\bin\bclanc.exe: *** fatal
error - could not load C:\Windows\system32\ws2_32.dll, Win32 error 998
Stack trace:
Frame Function Args
00289F34 6102740B (00289F34, 00000000, 00000000, 00000000)
0028A224 6102740B (6117AC40, 00008000, 00000000, 6117C997)
0028B254 61004B2B (6117C084, 0028B270, 00000000, 00000000)
0028B4A4 6100136E (61053A4A, 00000168, 00000002, 00000002)
The path is correct. I have no explanation why changing from a filename
to a pathname (and LoadLibrary to LoadLibraryW) would make any
difference, unless there's a race condition, say if LoadLibrary[W] is a
bit faster when you give it a full path. That's just speculation, of
course. But the intermittent nature of this crash (1 in 500ish
invocations) would also seem to point to a race condition of some kind.
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
