[DragonFlyBSD - Submit #1192] KKASSERTs in sys/kern/uipc_{msg,socket}.c are too strict

[sigttou]

Issue #1192 has been updated by sigttou.


I tested NFS-root with the current VKERNEL64: mounting, root traversal and
repeated file read/mmap checks pass. Restoring the old kva_p address-range
behavior reproduces the socket assertion in soopt_to_kbuf during the mount RPC,
which supports b9a7a2bd9a as the fix.

Native (non-vkernel) NFS-root boot also works: a client booted with its root on
an NFS export and completed its root checks with no panic or assertion.

The original report is about the vkernel case, so could that case be closed as
fixed? 

----------------------------------------
Submit #1192: KKASSERTs in sys/kern/uipc_{msg,socket}.c are too strict
http://bugs.dragonflybsd.org/issues/1192#change-14767

* Author: rumcic
* Status: New
* Priority: Normal
* Category: Kernel
* Target version: 6.4
----------------------------------------
The code for nfs root mounts does not use kmalloc-ed space for some vars, while
some KKASSERTs in the mentioned files only check for the supplied pointers if
they were kmalloc-ed or not.
The attached patch adds a check if the supplied pointers are not in user stack
(if they are in kernel stack) ... since kmalloc-ed space also resides in the
addresses not in user stack, maybe kva_p should be removed (is it used
anywhere else?)?
-- 
Regards,
Rumko

---Files--------------------------------
rootnfs.patch (2.63 KB)


-- 
You have received this notification because you have either subscribed to it, or are involved in it.
To change your notification preferences, please click here: http://bugs.dragonflybsd.org/my/account