Re: [PATCH -mm 7/8] user_ns: handle file sigio

2007-01-15 Thread Cedric Le Goater
[ ... ] > Rewriting the userns testcases right now. Clearly, in addition to > separately testing clone and unshare, I need to add a sigioperm check, > and have a separate set of testcases for CONFIG_USER_NS=n. Could we get rid of CONFIG_USER_NS ? It doesn't look that useful anyway, it just de

Re: [PATCH -mm 7/8] user_ns: handle file sigio

2007-01-15 Thread Serge E. Hallyn
Quoting Cedric Le Goater ([EMAIL PROTECTED]): > Serge E. Hallyn wrote: > > Quoting Andrew Morton ([EMAIL PROTECTED]): > >> On Thu, 4 Jan 2007 12:12:57 -0600 > >> "Serge E. Hallyn" <[EMAIL PROTECTED]> wrote: > >> > >>> A process in one user namespace could set a fowner and sigio on a file in > >>>

Re: [PATCH -mm 7/8] user_ns: handle file sigio

2007-01-15 Thread Cedric Le Goater
Serge E. Hallyn wrote: > Quoting Andrew Morton ([EMAIL PROTECTED]): >> On Thu, 4 Jan 2007 12:12:57 -0600 >> "Serge E. Hallyn" <[EMAIL PROTECTED]> wrote: >> >>> A process in one user namespace could set a fowner and sigio on a file in a >>> shared vfsmount, ending up killing a task in another user n

Re: [PATCH -mm 7/8] user_ns: handle file sigio

2007-01-14 Thread Serge E. Hallyn
Quoting Andrew Morton ([EMAIL PROTECTED]): > On Thu, 4 Jan 2007 12:12:57 -0600 > "Serge E. Hallyn" <[EMAIL PROTECTED]> wrote: > > > A process in one user namespace could set a fowner and sigio on a file in a > > shared vfsmount, ending up killing a task in another user namespace. > > > > Prevent t

Re: [PATCH -mm 7/8] user_ns: handle file sigio

2007-01-11 Thread Andrew Morton
On Thu, 4 Jan 2007 12:12:57 -0600 "Serge E. Hallyn" <[EMAIL PROTECTED]> wrote: > A process in one user namespace could set a fowner and sigio on a file in a > shared vfsmount, ending up killing a task in another user namespace. > > Prevent this by adding a user namespace pointer to the fown_struc