On Thu, 2009-10-29 at 14:08 +1100, Benjamin Herrenschmidt wrote:

On Wed, 2009-10-28 at 15:53 -0500, Nathan Fontenot wrote:

quoted

+	struct device_node *dn;
+	struct device_node *first_dn = NULL;
+	struct device_node *last_dn = NULL;
+	struct property *property;
+	struct property *last_property = NULL;
+	struct cc_workarea *ccwa;
+	int cc_token;
+	int rc;
+
+	cc_token = rtas_token("ibm,configure-connector");
+	if (cc_token == RTAS_UNKNOWN_SERVICE)
+		return NULL;
+
+	spin_lock(&workarea_lock);
+
+	ccwa = (struct cc_workarea *)&workarea[0];
+	ccwa->drc_index = drc_index;
+	ccwa->zero = 0;

Popping a free page with gfp (or just kmalloc'ing 4K) would avoid the
need for the lock too.

Not kmalloc -- the alignment of the buffer isn't guaranteed when
slub/slab debug is on, and iirc  the work area needs to be 4K-aligned.
__get_free_page should be fine, I think.