[Bug 281560] gve (4) uma deadlock during high tcp throughput

From: <bugzilla-noreply_at_freebsd.org>
Date: Wed, 18 Sep 2024 16:55:47 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=281560

Konstantin Belousov <kib@FreeBSD.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kib@FreeBSD.org

--- Comment #1 from Konstantin Belousov <kib@FreeBSD.org> ---
If you have the deadlock, there must be a resource acquisition loop, which
you should be able to unwind by looking at threads waiting for a resource,
then finding the owner of the resource, then  seeing what it is doing etc.

That said, do you run with INVARIANTS and WITNESS enabled?  If not, you must.
For instance, only INVARIANTS-enabled kernel would detect an attempt to
take a blockable lock inside smr section.  Similarly, WITNESS might directly
point to the resource loop for locks.

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