Re: [PATCH] add __GFP_ZERO to GFP_LEVEL_MASK

[Christoph Lameter <clameter@sgi.com>]

On Tue, 24 Jul 2007, Andrew Morton wrote:

> > __GFP_MOVABLE	The movability of a slab is determined by the
> > 		options specified at kmem_cache_create time. If this is
> > 		specified at kmalloc time then we will have some random
> > 		slabs movable and others not. 
> 
> Yes, they seem inappropriate.  Especially the first two.

The third one would randomize __GFP_MOVABLE allocs from the page allocator 
since one __GFP_MOVABLE alloc may allocate a slab that is then used for 
!__GFP_MOVABLE allocs.

Maybe something like this? Note that we may get into some churn here 
since slab allocations that any of these flags will BUG.



GFP_LEVEL_MASK: Remove __GFP_COLD, __GFP_COMP and __GFPMOVABLE

Add an explanation for the GFP_LEVEL_MASK and remove the flags
that should not be passed through derived allocators.

Signed-off-by: Christoph Lameter <clameter@sgi.com>

Index: linux-2.6/include/linux/gfp.h
===================================================================
--- linux-2.6.orig/include/linux/gfp.h	2007-07-24 12:31:04.000000000 -0700
+++ linux-2.6/include/linux/gfp.h	2007-07-24 12:32:50.000000000 -0700
@@ -53,12 +53,15 @@ struct vm_area_struct;
 #define __GFP_BITS_SHIFT 20	/* Room for 20 __GFP_FOO bits */
 #define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
 
-/* if you forget to add the bitmask here kernel will crash, period */
+/*
+ * GFP_LEVEL_MASK is used to filter out flags to be passed on to the
+ * page allocator in derived allocators such as slab allocators and
+ * vmalloc. It should not contain flags that are to be handled by the
+ * derived allocators themselves.
+ */
 #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \
-			__GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \
-			__GFP_NOFAIL|__GFP_NORETRY|__GFP_COMP| \
-			__GFP_NOMEMALLOC|__GFP_HARDWALL|__GFP_THISNODE| \
-			__GFP_MOVABLE)
+		__GFP_NOWARN|__GFP_REPEAT|__GFP_NOFAIL|__GFP_NORETRY| \
+		__GFP_NOMEMALLOC|__GFP_HARDWALL|__GFP_THISNODE| __GFP_MOVABLE)
 
 /* This equals 0, but use constants in case they ever change */
 #define GFP_NOWAIT	(GFP_ATOMIC & ~__GFP_HIGH)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/