Subject: Re: map user memory in kernel
To: Emmanuel Dreyfus <>
From: Chuck Silvers <>
List: tech-kern
Date: 11/05/2005 18:52:56
On Sat, Nov 05, 2005 at 11:41:49PM +0000, Emmanuel Dreyfus wrote:
> On Sat, Nov 05, 2005 at 12:09:59PM -0500, Nathan J. Williams wrote:
> > To make this work, the value to sleep on needs to be more persistent
> > than a physical address. Something like a hash of the vm_map and the
> > virtual address would do the trick for one process; to make this work
> > across processes, the right thing is probably the kernel address of a
> > lower-level backing object.
> An example for our kernel I could pick to finish Linux futex emulation?

I don't see any robust way to construct a unique sleep address per futex
with UVM.  for now, the best thing to do would be to use one global sleep
address for all futex sleepers.