On Mon, Jan 17, 2011 at 01:31:01PM -0800, Dmitry Torokhov wrote:

On Mon, Jan 17, 2011 at 02:29:20PM +0800, Thomas Chou wrote:

quoted

From: Walter Goossens <redacted>

For my education, why is this needed?

quoted

Signed-off-by: Walter Goossens <redacted>
Signed-off-by: Thomas Chou <redacted>
---
 drivers/input/serio/altera_ps2.c |   16 ++++++++++++++++
 1 files changed, 16 insertions(+), 0 deletions(-)

diff --git a/drivers/input/serio/altera_ps2.c b/drivers/input/serio/altera_ps2.c
index 7998560..93054a1 100644
--- a/drivers/input/serio/altera_ps2.c
+++ b/drivers/input/serio/altera_ps2.c

@@ -19,6 +19,9 @@
 #include <linux/platform_device.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#ifdef CONFIG_OF
+#include <linux/of.h>
+#endif

Does it have to be guarded?

Shouldn't need to be.

quoted

 
 #define DRV_NAME "altera_ps2"
 

@@ -173,6 +176,16 @@ static int __devexit altera_ps2_remove(struct platform_device *pdev)
 	return 0;
 }
 
+#ifdef CONFIG_OF
+static struct of_device_id altera_ps2_match[] = {

Not const?

quoted

+	{ 
+		.compatible = "altera,altera_ps2",
+	},
+	{},
+}
+MODULE_DEVICE_TABLE(of, altera_jtaguart_match);
+#endif /* CONFIG_OF */
+
 /*
  * Our device driver structure
  */

@@ -182,6 +195,9 @@ static struct platform_driver altera_ps2_driver = {
 	.driver	= {
 		.name	= DRV_NAME,
 		.owner	= THIS_MODULE,
+#ifdef CONFIG_OF
+		.of_match_table = altera_ps2_match,
+#endif

Can't we spare 4 bytes per driver and have .of_match_table always
present?

Yeah, probably.  I've been trying to avoid it, but I suppose that is
much ado about a tiny amount of data.  I'll craft a patch.

g.