Use work-stealing for load-balancing in the GC

  
New flag: "+RTS -qb" disables load-balancing in the parallel GC
(though this is subject to change, I think we will probably want to do
something more automatic before releasing this).

To get the "PARGC3" configuration described in the "Runtime support
for Multicore Haskell" paper, use "+RTS -qg0 -qb -RTS".

The main advantage of this is that it allows us to easily disable
load-balancing altogether, which turns out to be important in parallel
programs.  Maintaining locality is sometimes more important that
spreading the work out in parallel GC.  There is a side benefit in
that the parallel GC should have improved locality even when
load-balancing, because each processor prefers to take work from its
own queue before stealing from others.

1 files changed, 3 insertions, 1 deletions

diff --git a/rts/sm/GCUtils.h b/rts/sm/GCUtils.h
index 6948b2f1db..e71f4374fb 100644
--- a/rts/sm/GCUtils.h
+++ b/rts/sm/GCUtils.h
@@ -17,10 +17,12 @@ bdescr *allocBlock_sync(void);
 void    freeChain_sync(bdescr *bd);
 
 void    push_scanned_block   (bdescr *bd, step_workspace *ws);
-bdescr *grab_todo_block      (step_workspace *ws);
 StgPtr  todo_block_full      (nat size, step_workspace *ws);
 StgPtr  alloc_todo_block     (step_workspace *ws, nat size);
 
+bdescr *grab_local_todo_block  (step_workspace *ws);
+bdescr *steal_todo_block       (nat s);
+
 // Returns true if a block is partially full.  This predicate is used to try
 // to re-use partial blocks wherever possible, and to reduce wastage.
 // We might need to tweak the actual value.