agp/intel: Reduce extraneous PCI posting reads during init

Instead of doing a posting read after each GTT entry update, do a single one
at the end of the writes. This should reduce boot time a tiny amount by
avoiding a lot of extra uncached reads.

Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c
index f1fe749..81f0942 100644
--- a/drivers/char/agp/intel-agp.c
+++ b/drivers/char/agp/intel-agp.c
@@ -214,8 +214,8 @@
 	if (agp_bridge->driver->needs_scratch_page) {
 		for (i = 0; i < current_size->num_entries; i++) {
 			writel(agp_bridge->scratch_page, intel_private.registers+I810_PTE_BASE+(i*4));
-			readl(intel_private.registers+I810_PTE_BASE+(i*4));	/* PCI posting. */
 		}
+		readl(intel_private.registers+I810_PTE_BASE+((i-1)*4));	/* PCI posting. */
 	}
 	global_cache_flush();
 	return 0;
@@ -775,8 +775,8 @@
 	if (agp_bridge->driver->needs_scratch_page) {
 		for (i = intel_private.gtt_entries; i < current_size->num_entries; i++) {
 			writel(agp_bridge->scratch_page, intel_private.registers+I810_PTE_BASE+(i*4));
-			readl(intel_private.registers+I810_PTE_BASE+(i*4));	/* PCI Posting. */
 		}
+		readl(intel_private.registers+I810_PTE_BASE+((i-1)*4));	/* PCI Posting. */
 	}
 
 	global_cache_flush();
@@ -991,8 +991,8 @@
 	if (agp_bridge->driver->needs_scratch_page) {
 		for (i = intel_private.gtt_entries; i < current_size->num_entries; i++) {
 			writel(agp_bridge->scratch_page, intel_private.gtt+i);
-			readl(intel_private.gtt+i);	/* PCI Posting. */
 		}
+		readl(intel_private.gtt+i-1);	/* PCI Posting. */
 	}
 
 	global_cache_flush();