> While I haven't looked at your code closely enough, I'm pretty sure there is
> still a bug in there that involves the access of user data from the kernel in
> both sendsig and sigreturn.
> 
> Below are the diffs that correct this flaw. While the sigreturn function
> would do without the modification (as it only reads user data, and that
> it mapped into kernel space anyways), the sendsig code definitely needs
> the modification. Otherwise COW wouldn't work.
> 
I'll have a look after the other round of fixes. This looks like a different
problem thatn the one I'm going after though. Thanx.


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 John Brezak                    UUCP:     uunet!apollo.hp!brezak
 Hewlett Packard/Apollo         Internet: brezak@ch.hp.com
 300 Apollo Drive               Phone:    (508) 436-4915
 Chelmsford, Massachusetts      Fax:      (508) 436-5103


------------------------------------------------------------------------------