Subject: Re: lock bug in getnewvnode, or uvm_km_kmemalloc/uvm_map ?
To: Manuel Bouyer <bouyer@antioche.lip6.fr>
From: David Laight <david@l8s.co.uk>
List: tech-kern
Date: 11/21/2002 10:46:18
> It's just a question of being consistent with the interface: if we're calling
> uvm_map() with UVM_KMF_NOWAIT, then it should not sleep, whatever map it
> uses.
Maybe someone ought ot sanitise al the xxx_NOWAIT and xxx_CANSLEEP
(etc) options into a single type.
Then the value can be passed down from layer to layer safely.
David
--
David Laight: david@l8s.co.uk