I think it can be possible.
One month ago I notice that winedbg has problems when reading the process memory it tries to acess. Let me show the problem I encounter : - the windows program launched causes some error; - winedbg launches; - it tries to read a memory that is out of bounds; - a 2nd winedbg launches, trying to debug the first; - same error occurs (out of bound read); - a 3d winedbg launches, being successfull to read the memory of the 2nd.
Conclusion: Only wine code is used. (I defined all dll's built-in, and only advapi32 is loaded when using winedbg)
It's almost definitely a wine bug. "args[2]" becomes nil, get's deferenced, and once the SIGSEGV handler starts running, it never gets back to any native dll's -- that is to say, the infinite recursion happens
entirely within the wine codebase... SIGSEGV invokes wine handlers which then cause /another/ SIGSEGV, and so on.... shouldn't be possible, right?
___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com