[PATCH] uml: TLB operation batching

This adds VM op batching to skas0.  Rather than having a context switch to and
from the userspace stub for each address space change, we write a number of
operations to the stub data page and invoke a different stub which loops over
them and executes them all in one go.

The operations are stored as [ system call number, arg1, arg2, ... ] tuples.

The set is terminated by a system call number of 0.  Single operations, i.e.
page faults, are handled in the old way, since that is slightly more
efficient.

For a kernel build, a minority (~1/4) of the operations are part of a set.
These sets averaged ~100 in length, so for this quarter, the context switching
overhead is greatly reduced.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

1 files changed, 4 insertions, 1 deletions

diff --git a/arch/um/kernel/tt/tlb.c b/arch/um/kernel/tt/tlb.c
index 2eefb43bc9c2..16fc6a28882d 100644
--- a/arch/um/kernel/tt/tlb.c
+++ b/arch/um/kernel/tt/tlb.c
@@ -17,7 +17,8 @@
 #include "os.h"
 #include "tlb.h"
 
-static void do_ops(union mm_context *mmu, struct host_vm_op *ops, int last)
+static void *do_ops(union mm_context *mmu, struct host_vm_op *ops, int last,
+                    int finished, void *flush)
 {
         struct host_vm_op *op;
         int i;
@@ -45,6 +46,8 @@ static void do_ops(union mm_context *mmu, struct host_vm_op *ops, int last)
                         break;
                 }
         }
+
+        return NULL;
 }
 
 static void fix_range(struct mm_struct *mm, unsigned long start_addr,