tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

re: uarea swap-out

   Matt Thomas <> wrote:
   > If we are going to get rid of swappable uareas, then we should go all  
   > the way.
   > finally kill struct user (which just contains the pcb).  Move the pcb  
   > into the lwp as l_md.
   > The separate lwp pool should go away and the lwp should just be placed  
   > at the bottom of the uarea followed by the pcb.  l_addr should be  
   > renamed l_pcb.
   Right, these are more invasive. I will make a separate change set.
   Also, it was not in the patch, but I think we can remove pmap_collect().
   Since we would merge lwp_cache with uarea, I am not sure if we want to
   allocate lwp/uarea from lwp_create(). It would be good to improve failure
   case in fork1() (currently, handling lwp_create() failure in that path is
   complicated), and uvm_uarea_alloc() and uvm_uarea_free() can be useful.

i'm really curious how this affects platforms with very
few hardware pages, like acorn26.


Home | Main Index | Thread Index | Old Index