Re: DO_VOP/lockmgr related panic
:I agree, but for Emiel that could not even start X it would at least help him
:get some of his stuff done until someone come up with a proper fix.
:
:I will look more at it when I am done eating. Perhaps I can understand some vfs
:on the way.
:
:--
:Eirik Nygaard
:eirikn@kerneled.com
More eyes on the problem would definitely help, but I warn you it isn't
cut and dry. What is happening is that there is something (probably in
NFS) that is not leaving the vnode in the state the caller expects... but
the error is occuring *long* before the crash, possibly seconds to minutes
before the crash. It is only when the vnode is being reclaimed for reuse
in some later, unrelated operation that the panic occurs.
-Matt
Matthew Dillon
<dillon@backplane.com>
討論串 (同標題文章)
完整討論串 (本文為第 12 之 13 篇)：