Messages in this thread | | | Date | Thu, 1 Apr 2010 10:55:52 -0500 (CDT) | From | Christoph Lameter <> | Subject | Re: start_kernel(): bug: interrupts were enabled early |
| |
On Thu, 1 Apr 2010, David Howells wrote:
> > Can we provide a kmem_cache_create_early()? One that takes no locks and gets > cleaned up with the other __init stuff?
Sure. We can also check if we are early in boot in kmem_cache_create and just not take the semaphores (which are useless if we are single threaded).
| |