Messages in this thread | | | From | Matthew Dempsky <> | Date | Wed, 2 Apr 2014 15:37:33 -0700 | Subject | Re: [PATCH v3] ptrace: Fix fork event messages across pid namespaces |
| |
On Wed, Apr 2, 2014 at 2:58 PM, Matthew Dempsky <mdempsky@google.com> wrote: > Otherwise, it looks like (e.g.) the attempts to get the parent's pid > in fill_prstatus() and tomoyo_sys_getppid() are also theoretical races > of the same kind?
Doh, nevermind, those functions would potentially be use null "struct pid" pointers, not "struct pid_namespace" pointers; and pid_nr_ns() guards against pid being null.
Okay, I think I understand now why guarding against the task_parent_pid_ns() here is consistent with other RCU code. I'll send a revised patch that does that.
Also, it looks like "0" is the conventional value for unrepresentable pid_t values, so I'll change to do that too.
| |