https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=260591

Kubilay Kocak <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |crash, needs-qa
           See Also|                            |https://reviews.freebsd.org
                   |                            |/D31833,
                   |                            |https://reviews.freebsd.org
                   |                            |/D19375
              Flags|                            |maintainer-feedback?(imp@Fr
                   |                            |eeBSD.org),
                   |                            |maintainer-feedback?(mw@Fre
                   |                            |eBSD.org), mfc-stable13?,
                   |                            |mfc-stable12?
                 CC|                            |[email protected],
                   |                            |[email protected]
            Version|CURRENT                     |13.0-RELEASE
            Summary|[attach/detach] [panic]     |kern/subr_bus.c:
                   |insufficient multi-thread   |device_probe_and_{attach|de
                   |protection for              |tach}: Insufficient
                   |probe/attach/detach         |multi-thread protection
                   |                            |causes crash
             Status|New                         |Open

--- Comment #1 from Kubilay Kocak <[email protected]> ---
Some recent work [1] with Giant replacement in subr_bus and portentially
relatged work with respect to [2], which mentions "patch works around the real
issue, which is a locking deficiency of the busses.

Request feedback from those committers accordingly. 

[1] https://reviews.freebsd.org/D31833
[2] https://reviews.freebsd.org/D19375

^Triage: Set version to earliest supported and confirmed reproduction.

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to