Open raw-bin opened 7 years ago
There's this code from pfs_add_node() at sys/fs/pseudofs/pseudofs.c:108
:
#ifdef INVARIANTS
/* XXX no locking! */
if (pn->pn_type == pfstype_procdir)
for (iter = parent; iter != NULL; iter = iter->pn_parent)
KASSERT(iter->pn_type != pfstype_procdir,
("%s(): nested process directories", __func__));
I wonder whether that has a bearing on the problem ? The GENERIC conf enables INVARIANTS.
You can try GENERIC-NODEBUG
to avoid assertion panics. Of course they should be actually fixed though…
@myfreeweb Thanks. I commented out the code block referenced above and haven't had panics any more. Agree that a proper fix is preferable. I'll keep this open for the moment.
Hi.
I experience this panic frequently when undocking my Thinkpad T460s from it's Lenovo docking station. Prior to the undock I was using xrandr to get video output to an external monitor connected to the dock.
Relevant details are as under.
This is a link to a pic showing the backtrace.
Output of
uname -a
:FreeBSD gene 12.0-CURRENT FreeBSD 12.0-CURRENT #2 ff9c7fc1f1a(drm-next)-dirty: Fri Jul 14 21:29:44 BST 2017 robin@gene:/usr/obj/usr/src/sys/GENERIC amd64
The kernel is drm-next as of yesterday built with the bog standard GENERIC configuration. The dirty indication is because I have the following patch in place which was needed in order for suspend-resume to work correctly as reported here:
Please let me know if any other details would help. Any help/guidance graciously appreciated.
Cheers.